Monatsarchiv: Oktober 2013

Ein langer, aber sehr gut recherchierter Artikel über den Rennradfahrer, der auf der Suche nach einem neuen Streckenrekord auf Strava starb (auf englisch, via cycling.com)

 

No matter what else you might say about Kim Flint, you can’t claim that he was stupid. Flint was an electrical engineer at Nvidia, a $4 billion, 6,000-employee­ Silicon Valley firm (Forbesmagazine’s 2007 Company of the Year) known for making graphics-processing units, used in such products as Sony PlayStation. He managed a group responsible for signal integrity, which eliminates digital “noise” to ensure the quality of information transmitted and received by electronic devices. “Signal integrity is a niche field,” explains Flint’s former boss, Ting Ku. “It’s fairly small, exclusive, and rare. You’re working with digital signals in the gigahertz range—ones and zeros going really fast. Not many people understand it.” As a college undergraduate, Flint had been a Regents Scholar, the most prestigious aid award at the University of California; according to his parents, he had studied biology at UC Berkeley as a seventh grader and gotten the highest grade in the class. On his website, the Kimatorium, he ­posted papers he had written about his work—for example, “Effect of Bypass Capacitor Placement and Layer Switching.”

But Flint was not, says Ku, a typical engineer. “At the lunch table, he was one who did a lot of talking because he had interesting things to say about any topic. He was fun in that regard—he had experiences that not many people have.” In the 1990s, Flint worked for Gibson Guitar’s R&D lab for electronic instruments. During a ­sabbatical in 2003, he played jazz guitar while he taught himself signal integrity and tried to build a ­theremin—an instrument that can ­produce music without being touched by the musician, and a longtime staple of the sound tracks of ­science-fiction and suspense movies. In the music world, Flint was widely known as the creator of Looper’s Delight, a clearinghouse for information on the eccentric ­subculture of ­looping: repeating electronic samples that enable a solo player to create a textured, ­multilayered sound. Several of that discipline’s premier practitioners—Paul Drescher, Zoe Keating, Amy X. Newberg—were Flint’s friends.

With his girlfriend of 19 years, Violet Hefner—keeper of pet rats and rabbits, and the pink-haired curator of undented.com, a website devoted to Tori Amos—Flint lived in a rented warehouse on Adeline Street in West Oakland, a low-income, high-crime part of town near the eastern end of the Bay Bridge. One of their rooms was filled with Flint’s guitars and electronic devices; another housed a VW bug that never left the premises. Although he made good money, Flint commuted to work in an old Honda Civic, which he parked on the street at night with the windows open to prevent break-ins. “I drive up and down Interstate 880 a lot,” Flint disclosed on his website. “If you are ever in the fast lane and you see me behind you, please move out of the way. I’m tired of having to pass all you slow people on the right.”

Chris Aynesworth, an Oakland-based web designer, got to know Flint at a bar called Heinold’s First and Last Chance, a log-cabin-like saloon in Jack London Square, where a coterie of contrarians—a dog walker, a ceramic artist, a housepainter, a former Oakland port commissioner—convened on Friday nights “to get away from our lives and jobs.” Aynesworth found Flint “quietly brilliant, superconcentrated, cerebral. He had a quiet self-­confidence—he didn’t need to prove anything, he did things because he wanted to.” Neal Trembath, who shared a house with Flint in college and later worked with him at Gibson, says, “Kim was funny, sardonic, and sincere all at the same time. He was a clear, purposeful thinker who wanted­ others to think about things, not provide them with answers. He wanted people to question their own beliefs. He tried to make people’s lives better by challenging them in some way. He loaned ­money to homeless people on the street, but they had to pay him back. He even taught one of them, a guy named Will, computer skills for free so he could find a job.”

Still, Flint could be contentious—a fact that occasionally created friction at work. “His style was confrontational,” says Ku, who promoted Flint to manager after a few years at Nvidia. “He was an in-your-face type of person, and when something was not exactly right, he was very direct in pointing it out—not in private.” On his website, Flint gave vent to the edgier aspects of his personality. The message that greeted online visitors read, in part, “Your moral values are meaningless, artificial constructs.” The URL was annihilist.com, a nod to Flint’s nickname for himself, “Bitter Annihilist,” and, perhaps, also a description of his personal style: an almost-shaved skull backed by a bushy ponytail, punctuated by a 6-inch-long soul patch (or, as he called it, “goat beard”) dangling from his chin. He was also a vegetarian who, besides electrical engineering and computer science, had majored in peace and conflict studies in college.

WHEN FLINT STARTED riding in his late 30s, many of his friends were surprised. A small-framed guy of about 5-foot-9, he’d studied karate and played soccer and water polo in high school, but as an adult was never especially athletic. “Kim was always healthy,” says Trembath. “He did sit-ups every morning. But at midlife he decided he wanted to become superfit. He started drinking egg milk shakes and working out three hours a day. It was like Rocky.”

As part of his gym routine, Flint rode a stationary bike, bragging to friends that he pedaled with the resistance on its highest setting. Soon he got a Specialized road bike and embarked on a path familiar to anyone caught up in the conversion to two-wheeled religion: He got a handlebar computer, shaved his legs, and, four months before his 40th birthday, did a 104-mile ride. At first he retained his “annihilist” style—all black kit—but as he lost weight and gained muscle, he got rid of his trademark ponytail and let the hair grow in on top of his head. The soul patch remained, but he trimmed it down. Flint sought out the hardest climbs he could find into the Berkeley-Oakland hills: Centennial Drive, Claremont Avenue, Thornhill Drive, Hiller Highlands, Moeser Lane, Ascot Drive, El Toyonal, many of which contained grades of 10 to 15 percent. (“What’s wrong with you people, going up Butters?” he tweeted about a shady, 4 percent alternative to wide-open 8 percent Joaquin Miller Road. “Easy shit. Losers.”) Venturing farther from home, he tackled several of the Bay Area’s signature routes: Mount Hamilton in Santa Clara County, Mount Tamalpais in Marin, King Ridge in Sonoma. He entered the Mount Diablo Challenge, an 11-mile climb of a 3,800-foot peak in Contra Costa County, finishing 328th out of 1,000 (70th out of 151 in the 30 to 39 age group). Not surprisingly, he kept track of all his ride data and posted it online—a tendency that “really escalated when he got involved with Strava,” his partner, Violet, has said. “It became an obsession.”

Strava.com, the most successful, popular, and influential social fitness network to arise over the past half-dozen years, had launched soon after Flint started riding in 2009. Within a year, the cyclist would infamously come to epitomize its allure, value, and menace.

TRAVA WAS THE brainchild of Michael Horvath and Mark Gainey, who—stop me if you’ve heard this one before—were rowing teammates at Harvard in the late ’80s. Horvath, who spent the first six years of his life in Sweden (“Strava” means “strive” in Swedish), mainly grew up in Hingham, Massachusetts, south of Boston, where he was “the kind of kid who would hang around at the bike shop.” In high school he played soccer and tennis, but he says, “It wasn’t until I got to college and joined the crew team that I realized that I had the capacity for winning anything.”

Rowers, Horvath explains, train all year long for a half-dozen races in the spring. “To stay motivated, it’s really important to have teammates you’re training with day in and day out, who drive you and push you to achieve your peak in performance. That’s almost what being on the crew team is all about, as opposed to ­racing against other crews. And it’s hard to re-create when you leave.”

After college Horvath became a triathlete. He trained alone mostly, working around a growing family and a job teaching economics at Stanford. Meanwhile, his old teammate Gainey was working at a venture-capital company in nearby Silicon Valley, where the World Wide Web was beginning to infiltrate people’s personal lives. “Mark would come to my office and pitch ideas for different start-up companies,” Horvath remembers. Together, the two ex-rowers hatched the notion of creating a kind of online locker room. “What we wanted to do is connect you with other people so you don’t feel so alone, even when you’re training alone,” says Horvath. “If you can see yourself relative to other folks and you know your workout is going to be exposed to them after you’re done, it’s going to motivate you.”

They floated the idea to various website developers, none of whom took them up on it. “In 1996 the technology was very rudimentary—everyone said, ‘No way we can build that for you.’ But some developers suggested another problem to solve: Customers wanted to talk to companies through e-mail, but didn’t have any ­systems to handle that flow of communication.” Horvath and Gainey started Kana Software, the first enterprise e-mail management system, and it went public three years later. After cashing out of Kana, ­Horvath moved to New Hampshire to teach entrepreneurship at Dartmouth’s Tuck School of Business. Gainey stayed in Silicon Valley, where the tools for building an online locker room were rapidly improving.

“In our original conception, Strava was just going to be manually entered workouts,” Horvath says. “GPS wasn’t even part of it—it used to be very expensive to get a Garmin watch to track your workout, and very poor quality. It was really just an approximation of where your location was—you couldn’t even tell, ‘Am I on that street or this street?’ But during the Clinton administration, the government increased the accuracy level for public use of GPS from 50 meters to 3—and the price point had come down into the three-hundred-dollar range.”

Moreover, with the advent of Twitter and Facebook, the practice of sharing information online had gone public in a big way. And social fitness networks such as Garmin Connect and Map My Ride were luring legions of cyclists to upload their distance, speed, elevation gain, heart rate, pedaling cadence, and power output. Clearly, Horvath says, “The time was right to try our idea again.” They focused on cycling (as opposed to running, which they’ve since added) because bike riders were already widely using GPS and power meters. The difference between Strava and its competitors—most of which consisted merely of elaborate storehouses for personal data—was the public comparison of one’s own performance with others.

“It might be people you know, or people you’ve initiated a relationship with and are following,” Horvath says. “If someone is following you, they’ll see the information you’ve uploaded. If you look at my profile, for example, you’ll see yourself compared to me on average rides per week for the last four weeks, total number of miles ridden, and hours on the bike this year. It gives you an ability to see if a person is extremely active, about as active, or less active than you are—if they’re compatible with you if you go for a bike ride.”

The other original, even more crucial component (“the center of the experience,” in Horvath’s words) is Strava’s tracking of your performance on segments of popular routes. “It could be a climb,” Horvath explains, “or it could be a sprint. We compare your time on that segment to the times of everyone who’s ridden it before and uploaded it to Strava.” Thus, if you bicycle up Tunnel Road in Berkeley, Strava automatically ranks you on a leaderboard among (as of this writing) 3,682 other members who have done it, with the first-place finisher designated the King of the Mountain, or KOM (plus the Queen of the Mountain, for women), and with runners-up identified by numbered trophies. Even if you’re nowhere near the top, you win a PR icon if you achieve a personal record.

“It creates a sense of motivation,” says Horvath. “Can I get better relative to my own fastest time, or relative to other people­ I know? There’s a sense of friendly­ competition among the people you’re connected with.”

Membership is free, but for a fee you can upgrade to Premium status, which allows you to filter your data. “If I’m ranked four-hundredth out of twelve hundred, it may not be so relevant,” says Horvath, “but I can also ask: How do I compare to the people I’m following? Or: How do I compare to people in my age group? Or my weight class?” Since you enter such personal information (as well as details such as the weight of your bike) in your user profile when you sign up, even if you don’t use a power meter Strava can calculate how much effort it took to haul yourself from Point A to Point B at X miles per hour. If you have a heart-rate monitor, it can also display a “suffer score.”

The company doesn’t release membership numbers, but says that participation grew by a factor of 15 to 20 in its first year and 8 to 10 in each of the three years that have followed. With apps available for iPhone and Android, as well as the ability to import information from Garmin and other GPS units, membership extends to 221 countries and seven continents. It’s no exaggeration to say that, in just four years of existence, Strava has transformed the cycling world, winning praise and devotion for its motivational capability, how it aids and simplifies performance analysis, its usefulness for discovering new routes, and help with forging new friendships. Meanwhile, the experience of competing, formerly obtainable only by showing up on a starting line with other flesh-and-blood riders, can be had in private.

Like any other game-changing innovation, this one has its critics and pitfalls. Chief among these is, indeed, the charge of gamification: Strava-besotted cyclists engage in personal time trials when­ever they mount their bikes, and with so many users familiar with the locations of popular segments, formerly civil training rides now explode at predictable—if intangible—points. (“Hundreds of invisible city-limits signs all over the place!” a user named Kristofer Wagner exulted. “Pure awesomeness!”) Like teammates in a traditional race, riders sometimes work together for the sake of snagging segment wins, and individuals plot strategies to optimize their Strava numbers—for example, purposely starting a climb off the rear of a group then advancing to the middle or front so that, even if the group summits together, the late starter has the fastest time by the end (a practice known as Stravasniping). Strava has enabled mountain bikers to easily discover and publicize illegal trails and has led to accusations that hikers and equestrians are being endangered by gonzo downhillers in pursuit of records. So deeply has Strava become embedded in the sport that, in cycling parlance, its name has now become a verb—and overdependence on it has, correspondingly, inspired some colorful nouns, including “Stravaddict” and ­“Stravasshole”—the latter denoting, according to Kris Thompson on the Boulder/Denver area website 303cycling.com, “an individual Strava-ing at the expense of common courtesy.”

Wayne Lumpkin, owner of Spot Brand Bicycles in Golden, Colorado, and a longtime icon of the sport, summed up a common sentiment when he told Bicycle Retailer, “People call it social media, but I call it anti-social media.”

FOR AN ANALYTICAL person like Kim Flint, Strava proved irresistible. He became such an avid user that he got in touch with Strava’s engineers, helping them work out bugs in their system for establishing segments. Over time, he created 39 segments himself. One might be tempted to suggest that, as a devoted looper, he embraced Strava because he enjoyed playing alone—but in fact, as the company foresaw, he also reached out to local members whose times were close to his.

“Hey, you’re on Strava right?” Flint tweeted to an Oakland cyclist named Steve Shores in October 2009. “The guy who beats me by 30 seconds on every damn climb in the East Bay?”

“Yeah but you are close,” Shores replied.

“We should do a ride sometime,” Flint answered. “I’m having a lot of fun with Strava, they’re doing a good job on that.”

Like compatible online daters, Shores and Flint soon started riding together in real life. In early December they climbed 4,000-foot Mount Hamilton amid snow flurries, an achievement that Shores celebrated by tweeting: “Welcome to the top 5 club! You did an awesome ride today!”

“Yeah, you too! Are you looking at the leaderboard? I’m #1 best climb of the week?”

After you reach the top of a climb, of course, you get to go down—a prospect in which Flint and Shores found further communion. On his Twitter page, Shores described himself as an “adrenaline needer,” and soon the two riders were spurring each other to downhill KOMs.

“Added Spruce descent to strava.com, so ­@steveshores can do his maniac downhill thing,” Flint tweeted in January 2010, referring to a popular Berkeley hill that happens to contain four stop signs. But despite this building brinkmanship, Flint’s friends say he wasn’t reckless.

“Kim was pretty cautious and calculating by nature,” says Trembath. “He was more a thrill observer than thrill seeker. But he wanted to challenge social conventions, and he was attracted to edginess. The hubris he seemed to develop was exactly the sort of thing he would’ve made fun of and challenged in somebody else when he was younger.­ His Strava bravado probably started out as ironic and developed into a genuine desire. I figured he’d burn through it, but underneath it I think he was really proud of his ability.”

IN THE SPRING of 2010, Flint and Shores started riding with a group of Berkeley cyclists they’d met through Strava. All were about 40 years old; one, Patrick Gordis, was a seasoned veteran who’d won his first race in 1976 at age 10 and whose brother Kent has produced network-TV coverage for six Tours de France and coauthored Greg LeMond’s Complete Book of Cycling. Two of the others, a pair of El Cerrito neighbors named Steve Zavestoski and Matt McHugh, had, like Flint, been riding for only a couple of years, but were rivals on Strava.

The first ride they did with Flint was a 100-mile epic to the far side of Mount Diablo and back. The centerpiece was Morgan Territory Road, a remote, twisty one-lane that ascends gradually through an oak forest then drops 1,400 feet in 4 miles through open ranch country. Gordis noticed that Flint wasn’t comfortable in a paceline: “He was hanging off the back and doing his own thing. Several times I thought we’d dropped him, but at stoplights he’d catch up.” On the descent, however, Flint got ­Gordis’s attention in a new way. “I’m a medium descender,” Gordis says. “I’m capable of going fast if I force myself, but Kim and Steve Shores were way ahead of me. They were both really into the Strava downhill segment. They just absolutely took off.”

Over the next few weeks, the group did more de facto centuries, including one that Gordis calls the Grit Route: south through the flatlands of Oakland into the industrial suburbs of Fremont, then east on Niles Canyon Road into the hills. “When we got to the turn for Niles Canyon, somebody told me it was a Strava segment,” Gordis says. “We were going 23 or 24 miles an hour for 6 miles. Again, Kim was riding off the back, not even wheel sucking. We didn’t distance him; he was really strong, so he was keeping pace but riding very inefficiently. He was also pushing too big a gear—he was always in his smallest cog, or one down. My sense was that he hadn’t had a lot of feedback or mentorship, so when we came to a rest stop, I told him that he should learn to draft more efficiently and maintain a higher cadence.”

Perhaps because of his results on these rides—and the growing popularity of Strava—Flint soon narrowed his focus. “He thought the climbs were getting too competitive,” says Matt McHugh, “so he’d go after descents.”

As Gordis says, “Downhills are the easiest way to get high up on Strava. You don’t need a lot of fitness, so you can leapfrog the process. Basically, it’s a game of chicken.”

ON JUNE 6, Flint did a 75-mile ride in the East Bay with 7,349 feet of climbing. “Hard ride today,” he tweeted. “Set new personal records—Centennial, 3 Bears, some others. Even a KOM on south gate descent!”

By “south gate,” Flint meant South Park Drive, the fastest descent in the Berkeley hills—a 1.4-mile, 700-foot parachute drop (9.6 percent average grade, with pitches as high as 20 percent) that contains a series of blind curves. Located in Tilden Regional Park, it’s punctuated by trailheads and picnic sites, and, though closed to car traffic in winter, is commonly crisscrossed by wildlife and popular with dog walkers. Flint’s Strava readings for June 6 show his average speed on South Park Drive to be 39.9 mph, his average heart rate 153, and his power 274 watts; even going downhill, he was putting out a lot of effort, enabling him to hit his fastest speed ever. “49.3 mph, on a bike,” he subsequently tweeted, attaching a photo of his Garmin screen as proof. “How I find religion on Sunday morning.”

“In a 30 mph zone,” commented a Strava member. “Sounds pretty dangerous. RIP dude.”

ON JUNE 13, Flint, Gordis, and friends did a ride that took them to Palomares Road, which climbs for 5 miles then plunges 600 feet in a mile before assuming a gentler profile. Again, the group lost quickly sight of Flint as he rocketed downhill. After reentering Oakland, they descended Joaquin Miller Road, a plummeting four-lane speedway lined with parked cars and intersected by multiple cross streets. “Kim disappeared at a terrifying rate of speed,” Gordis told the group in an e-mail. “I was afraid I’d find him in a body bag at the bottom.”

After uploading his ride data, Flint was incredulous that he’d failed to win a downhill KOM. The first-place rider on Palomares was a top local racer of 20 years’ standing: Larry Nolan, winner of 42 national and 112 state elite and master’s road and track championships, holder of one world and two national records, and, as of that year, directeur sportif of Team Specialized, the company that made Flint’s bike. None of that seemed to make an impression on Flint. “It’s not right to see a descent in the East Bay without SteveS or me at the top!” he wrote to Gordis.

“He had no idea what he was up against,” Gordis says. “I was worried about him.”

ON SATURDAY, JUNE 19, Flint was supposed to meet Chris Aynesworth to work through a ­computer problem. “He e-mailed me and said he had to run a few errands first,” Aynesworth remembers. “He was going to take a ­leisurely ride by himself to get coffee and mail from his post-office box in Berkeley.” That was Flint’s regular Saturday-morning ritual, but on this day he was distracted by something that had occurred earlier in the week. On Tuesday, a rider named Pan Thomakos—who happened to be a Strava software engineer—had taken the South Park Drive KOM, beating Flint’s record by four seconds. According to Flint’s Garmin, at 2 p.m. on Saturday, he was at the top of South Park Drive.

At 4:45, Flint’s girlfriend Violet posted a panicked message on Twitter: “Kim’s been hit by a car. I’m waiting for a ride to the hospital. OMG OMG OMG OMG OMG. Please please PRAY.”

Aynesworth picked up Hefner and drove her to John Muir Hospital in Walnut Creek, where the cyclist had been airlifted by ambulance helicopter. Flint’s parents, who live in nearby Danville, were already there. “At the hospital,” Aynesworth says, “I overheard a woman on a walkie-talkie. She said Kim was going around a curve, lost control, tried to stop, and couldn’t. He rammed an SUV, flew 40 feet in the air, and crushed his chest. His heart and lungs were in bad shape—he tried to talk, but only said, ‘I can’t breathe. I can’t breathe.’ He said his name was Kevin, which is his brother’s name.”

The place where Flint crashed is only a quarter of a mile from the top of South Park—a sharp right downhill curve where a rider intent on taking it at high speed would have to cross into the oncoming lane to cut back through the apex. Adjacent to a picnic parking area on the left side of the road, it’s the lower part of a sweeping S-curve that follows a short rise in the plunging hill—an upturn so short that you don’t even have to resume pedaling, but which briefly blocks your view of the road ahead.

“It’s the only tricky turn on the whole descent,” says Steve Zavestoski. “Kim was a very analytical person; he knew every second he had to shave off.” Adds Aynesworth: “He knew the corners. He was just trying to go as fast as humanly possible. His father later told me that his bike was in perfect condition—his body absorbed all the impact.”

The SUV driver was a woman who had her young daughter in the backseat. They were heading uphill in their own lane when Flint came around the corner; according to park police, Flint braked so hard that he nearly flipped, fishtailed into the side of the vehicle, and ricocheted off.

“His skid marks went over the center line,” says Aynesworth, who visited the site two days later. “We found part of an eyepiece of his sunglasses with blood on it. It was horrible.”

Flint had suffered severe injuries to his head, neck, spine, ribs, femur, liver, diaphragm, and right lung. “If he hadn’t died, he probably would have been paralyzed,” says Neal Trembath.

“I saw his parents when they were told that he’d passed away,” says Aynesworth. “I had to hold Mrs. Flint up—she couldn’t stand. Mr. Flint was very angry. He just said, ‘Let’s get out of here.’”

FOR THE FIRST few days after news of Flint’s death was reported in the San Francisco Chronicle, most of the online cycling commentary focused on the dangers of descending South Park Drive. Many riders reported going even faster there than Flint had—some apparently bragging, but some expressing an attitude of chastened self-reform, citing a sense of responsibility to their families.

It took about a week for the Strava connection to surface. In late June and early July, the independent news site Berkeleyside published a pair of articles entitled “Did attempt to set speed record cause cyclist’s death?” and the even more suggestive “Did bicycle website contribute to cyclist’s death?” Another piece by the same writer (Frances Dinkelspiel) in theNew York Times mentioned Flint’s “obsession with Strava,” a perspective confirmed by Violet Hefner in the UC Berkeley student newspaper The Daily Californian. Hefner stopped short of blaming the website: “They’re trying to do something to help motivate people. I think that’s a good thing.” But she said it fed his urge to push his limits.

Not everyone accepted the characterization of the incident as an act of Stravacide. “I would not say that Strava attracts the speed addicts who are really looking for ways to go out and post KOM rankings,” Steve Zavestoski told Berkeleyside. “Strava attracts people who want to be better at cycling, better on the climbs, better on the distance. It just so happens you can’t go up a hill without going down a hill.”

Of course, Flint’s attitude toward descending wasn’t exactly one of reluctant resignation. His fears for Flint confirmed, Patrick Gordis told the Times, “Strava should not, even inadvertently, enable this kind of downhill dive bombing­ for time on open roads.”

Eleven days after Flint’s death, Strava announced that it would enable any member who had ridden a segment to flag it as ­hazardous, a designation that would eliminate its leaderboard. The South Park segment subsequently vanished—flagged by none other than CEO Michael Horvath. But the segment soon resurfaced, a loophole that can be achieved when users slightly alter the start or end point. Other descents (on Mount Diablo, for example) started showing up with names like “We Aren’t Going to Flag This Segment Because We Aren’t Sissies Wearing Dresses.”

CLEARLY, STRAVA TAPS into some deep part of the brain—not just ancient competitive urges, but modern electronic ones. In search of a professional perspective, I contacted the Association for Applied Sport Psychology in Indianapolis, which referred me to Kristen Dieffenbach, an assistant professor of athletic coaching education at West Virginia University.

A PhD in exercise science with an emphasis on sports psychology, Dieffenbach—also a licensed elite-level USA Cycling and Level II endurance track and field coach, and an ultraendurance racer herself—says, “People were doing back-alley races long before Strava. A lot of ­people thrive on competition. The problem is when it’s out of balance—when competition, as opposed to the experience, becomes the emphasis. As a clinical psychologist, that’s where I get concerned.”

Dieffenbach has worked with all sorts of professional, recreational, college, and high school athletes. Cyclists, she observes, “have more data at their fingertips than almost any other sport—all these external, extrinsic sources for how they’re performing, which can make it easy for people to get in over their heads a little sooner than they’re ready for. It’s so easy to get caught up in that stuff. I’m not a neuroscientist, but I’ve done a good bit of reading on how our culture and electronics are processing information differently these days. You can’t blame the electronics—it’s not a causal relationship, but we live in a sitcom/solve-it-in-27-minutes culture, and when someone is high-performance driven, the immediate feedback of devices serves that need. People spend the whole time staring at their device trying to stay at 250 watts, and they lose touch with the intrinsic feeling of the ride. They can’t even tell me: What did it feel like?”

When Dieffenbach reads about “how our culture and electronics are processing information differently these days,” she could be referencing the work of Peter Whybrow. Director of the Semel Institute for Neuroscience and Human Behavior at UCLA, and author of three books on the subject, Whybrow recently attracted attention for calling the computer “electronic cocaine”—a reference to the fact that, biochemically speaking, the brains of Internet junkies resemble those of traditional drug and alcohol addicts.

Most discussions of electronic addiction center on dopamine, the “feel-good” chemical the brain releases when we achieve a goal or anticipate positive news. Originally central to our survival (as a reward for things like finding food or having sex), today dopamine can be triggered by such diverse stimuli as music, chocolate, nicotine, gambling, getting e-mail, or uploading data to Strava.

Whybrow says the dopamine cycle—which is always involved with addiction—is driven by three things: curiosity, self-interest, and social competition. Strava, he notes, combines all three. “If you’ve narrowed the pleasure you get from a bike ride to a pure abstraction of how long it takes—if you’re not interested in where you rode, how you’re riding, or who you’re riding with—that’s exactly what happens with addiction,” says Whybrow. Addicts are totally preoccupied with one small element: ‘Where am I going to get my next shot?’”

In other words, electronic devices (and online social networks) are addictive because they connect with our most ancient instincts. “The primitive, central part of the brain is millions of years old,” Whybrow says. “We’re competitive animals because of the basic need to survive—we had to be attentive not only to novelty, but to getting there first to eat the fruit off the tree. Self-control comes from a different part of the brain—a more refined part, the prefrontal cortex or ‘executive brain.’”

Basically, this is the part that applies the brakes and saves some of the food instead of gobbling it all up right away. The need for self-control, Whybrow says, hinges on “whether­ the environment is with you or against you.” In cycling, for example, “If you’re going uphill, gravity is a natural constraint—you don’t have to use the prefrontal cortex. But when you’re going downhill, gravity is working with the primitive part of your brain to kill you.” Normally, the prefrontal cortex intervenes before we succumb to catastrophe—but in extreme conditions of competition, abundance, or addiction, concludes Whybrow, decision making “is reduced to a very small part of the brain.”

All of which might explain what, two years after Kim Flint got killed, Chris Bucchere did in March 2012.

LIKE FLINT, BUCCHERE—who was 36 at the time and a husband, father, software engineer, gourmet chef, and mentor for disadvantaged kids—was a Strava enthusiast. (“How long before winning a Strava segment = getting a free coffee at Velo Rouge?” he mused on San Francisco’s Mission Cycling Club news group on March 25, 2012.) But Bucchere was an experienced cyclist—he’d mountain biked as a teenager in the East Bay hills and joined the Stanford cycling team in college. He didn’t belong to the 135-­member Mission club, but one of his regular rides was that group’s popular Headlands Raid, a spectacular dawn jaunt across the Golden Gate Bridge.

At 6 a.m. on March 29, Bucchere and a friend named Stephan Morais set out on that route. They turned around at Hawk Hill, with its sweeping views of the city and coast to the south, then returned via the Presidio and Golden Gate Park, parting company below the Haight Ashbury ­district, near the top of a hill where Divisadero Street turns into Castro and descends toward ­Market—a Strava segment then known as the Castro Bomb.

At about 8 a.m., while driving a Honda SUV south on Castro Street, a local resident (and frequent bike rider) named Nathan Pollak found himself leapfrogging down the hill with a cyclist who was riding a Stradalli Trebisacce racing bike. At stop signs and red lights where Pollak slowed his automobile to a halt, the cyclist flew through without constraint. In the last block before Market Street—a major intersection and site of numerous transit stops and rush-hour foot and vehicle traffic—the grade steepened, enabling the cyclist to pick up yet more speed.

“The light turned red before the cyclist hit the intersection,” Pollak would later testify. “[But] he continued through. My jaw dropped…but to be honest, I wasn’t surprised. He crouched down to push his body weight forward and intentionally accelerated.”

Later that day, a person who identified ­himself as Bucchere on the Mission Cycling message board offered his perspective on what happened next. “The light turned ­yellow as I was approaching the intersection, but I was already way too committed to stop,” he wrote. “The light turned red as I was cruising through the middle of the intersection and then, almost instantly, the southern crosswalk on Market and Castro filled up with people coming from both directions.”

One of those people was Angelo Cilia, who later said that, entering the crosswalk after the walk signal came on, he heard a voice yell, “Hey!” followed by “a bone-crunching sound.” Another pedestrian on her way to work, a young woman named Wen-Chih Yu, said she glimpsed a bike going “incredibly fast” before it ran into an elderly pedestrian crossing from the other side. “The old man didn’t have a chance to put his arms out or anything,” Yu said. “He flew and skidded into the pavement with his face.”

“In a nutshell, blammo” was how Bucchere described the impact on the message board. “I couldn’t see a line through the crowd and I couldn’t stop, so I laid it down and just plowed through the crowded crosswalk in the least-populated place I could find. I don’t remember the next five minutes but when I came to, I was in a neck brace being loaded into an ambulance. I remember seeing a RIVER of blood on the asphalt, but it wasn’t mine. Apparently I hit a 71-year-old male pedestrian and he ended up in the ICU with pretty serious head injuries. I really hope he ends up OK…. Anyway, other than a stiff neck, a sore jaw/TMJ, a few bruises and some raspberries, I’m totally fine. I got discharged from the hospital during the lunch hour. The guy I hit was not as fortunate. I really hope he makes it. The cops took my bike. Hopefully they’ll give it back.”

Sutchi Hui—the 71-year-old pedestrian who had been on his way to Kaiser Permanente with his wife to pick up a pair of eyeglasses—was not okay. He died four days after being admitted to the hospital. Subsequently, Bucchere was charged with felony vehicular manslaughter, carrying a potential penalty of six years in prison. “Mr. Hui was a husband and father,” said the San Francisco district attorney, George Gascon, “and he was killed because of a bicyclist’s need for speed.”

At ensuing Internet speeds, Bucchere became a cycling pariah. The San Francisco Bicycle Coalition condemned him; the Mission Cycling Club disavowed association with him; through various online avenues he was the subject of death threats and, more commonly, character assessments such as “giant asshole” and “real douchebag.”

And Strava again found itself connected to a controversial fatality. “As a Strava user, my first thought when I saw that he was using Strava was that he was trying to post the best time,” One Thousand and One wrote on ­metafilter.com. “That stretch of Castro is, indeed, a marked segment, which is absolutely fucking stupid and likely encouraged in some small way his reckless behavior.”

[ Related: Cycling Lawyer Bob Mionske on Bucchere ]

“Did he win that free coffee?” Ernest wondered on sfcitizen.com.

“Strava might get sued,” said sfcitizen. “We’ll see.”

Strava did get sued—but not by the family of Sutchi Hui. The plaintiffs were the parents of Kim Flint, who, convinced by the Bucchere case that “enough is enough,” filed a suit for wrongful death on the last day before the two-year statute of limitations expired.

IN THEIR LEGAL filing, the Flints sought to hold Strava accountable for general negligence. They said that the company “breached their duty of care by: (1) failing to warn cyclists competing in KOM challenge that the road conditions were not suited for racing and that it was unreasonably dangerous given those conditions; (2) failing to take adequate measures to ensure the KOM challenges took place on safe courses; and (3) encouraging dangerous behavior.” They concluded that, had Strava taken the necessary measures, “Kim Flint Jr. would not have died as he did.”

“The death of Kim Flint was a tragic ­accident, and we expressed our sincere condolences when it occurred in 2010,” Mark Riedy, a spokesman for Strava (and a former editor at Bicycling), said in an official statement, “but we will defend the company vigorously through the legal process ahead.”

I met the Flints in August 2012 near Jack London Square—just down the street from the Oakland bar where Kim had convened with friends on Friday nights—in the office of their attorneys, Richard Meier and Susan Kang. William Flint Sr., a white-haired former career naval officer (now a certified financial planner who also goes by the nickname Kim), was, like his son, a lean and slightly built but pugnacious guy. By contrast, his wife, Kathleen, was very quiet—rarely speaking at all, and then only to him in a whisper.

“She died the day my son died,” Flint Sr. explained, adding that his wife hadn’t been able to work in the two years since. “My contention is that if [Kim Jr.] had never discovered Strava, he never would have gotten involved in doing dangerous descents on South Park or anywhere else.”

His head cocked at a combative angle, Flint Sr. declared: “Strava was responsible for the death of my son.”

Kang sounded a more temperate note: “We’re not saying he didn’t assume any risk. It’s a shared responsibility. But for Strava to say they have no part, they’re lying to themselves.” She said the aim of the lawsuit wasn’t financial compensation, but safety—for cyclists and also for the drivers, pedestrians, and hikers whom she said Strava endangers with a new and uncontrolled kind of racing.

A young, fit-looking woman (and former clerk for the state bar association and San Francisco district attorney’s office), Kang said she’d been a competitive runner since the age of 14. “We used to sign up and get a number,” she reminisced. “The [entry] fee ensured that the race director would be responsible for safety—streets would be roped off and regulations would be followed. But now we have a different type of racing—an open-ended system of ‘invisible’ race promotion. To be relevant and competitive, you have to run stop signs and break the speed limit. In what other capacity is this allowable?”

Meier said Strava has “a responsibility to temper their outreach,” and Flint Sr. suggested­ this could be done by eliminating leaderboards for downhill segments. “They might lose popularity,” he said, “but they’d save lives.” He added that, since his son had died, 21 ­Strava members had gone faster than Kim down South Park Drive.

Flint Sr.’s insistence that Strava encouraged his son’s conduct was based in part on his belief that the company had alerted Kim when the South Park record was broken. He’s mistaken. In 2012, almost two years after Flint’s death, Strava did start sending messages announcing, “Uh oh! [So and so] just stole your KOM….Better get out there and show them who’s boss!” (The company has since changed the alert to conclude by exhorting the vanquished to “Get out there, be safe and have fun!” And the current notice is sent only for segments with a grade of 0 percent or more—flats or ascents.)

Alerts aside, the fact that Flint’s fateful KOM was taken from him by a Strava employee was provocative. Thus, in April 2013, ­Meier and Kang spoke with Pan Thomakos in a pretrial deposition. They learned, among other things, that Thomakos started working at Strava in August 2009; that he owns shares in the company; that as an employee he had been given a GPS device, but no one had asked him to ride his bike; and that since Flint’s death, he had stopped using Strava to track his own rides.

Among the many things Thomakos said he couldn’t remember (in response to questions) were whether he had ever ridden his bike on company time; whether he had ever held a KOM title; whether anyone at Strava had ever asked him about any specific segments; whether, in going through the entire history of his e-mail inbox, he’d found any messages referring to South Park Drive; whether that road had a posted speed limit; and whether he had ever gone faster than 30 miles per hour on it.

Thomakos did say that when Strava posts users’ times, it doesn’t account for safety, danger, stop signs, speed limits, or the fact that, to beat some KOM times, a rider would have to violate the law.

MANY SUGGESTIONS HAVE been offered for how Strava could make the downhill ­segments safer. One is for the website to ascertain the legal speed limit on every segment and not post any user time that exceeds it. Another is to appoint regional ambassadors who set limits on KOMs, or simply prohibit segments that contain a net loss of elevation or traverse an intersection.

The day after I met the Flints, I put these ideas to Strava’s CEO Horvath. At the time, Strava was in 160 countries instead of the 221 it is now; referencing the former number, he said, “We cannot monitor what’s going on at the level of local riders—they have to do that. But we’re starting to understand that we need to be proactive in getting that message out.” In June 2012, Horvath created a new page on Strava entitled “Stand With Us,” enumerating five guideposts: “We know the rules (‘Cycling, running, and swimming are inherently dangerous’); We rest (‘We don’t burn ourselves out’); We kudo sportsmanship (‘We are ­courteous and treat others with respect’); We think ahead (‘If we don’t want everyone to know what we’re up to, we take the necessary privacy precautions before we upload’); We’ve got each other’s backs (‘Keep things safe for everyone by looking out for potentially dangerous situations and flagging segments as hazardous’).”

When it comes to complying with speed limits, Horvath told me, “That’s not something we’re monitoring or tracking or have any data insight into.” Downhill segments? “Descending is something we all do—it’s part of cycling, it’s part of racing. Is [the idea] that the world can’t be safe with descending segments? [That’s like] saying we shouldn’t be doing any descents on our bicycle—that we should get off and walk down. I don’t think that’s true. You need to take responsibility and use common sense, same as when you’re crossing an intersection or rolling through a stop sign. I’m counting on the cyclists to be responsible for their safety, and giving them a tool to do something about it—anybody who’s ridden a segment has the ability to shut the leaderboard down if they’re concerned that it’s no longer safe.”

STRAVA WAS REPRESENTED in the lawsuit by the firm of O’Melveny & Myers—an 800-attorney, Los Angeles-headquartered­ company with offices spanning from Brussels to Beijing, and clients that have included Goldman Sachs, Exxon Mobil, Bank of America, and former Enron CEO Jeffrey Skilling. Within a few months, they filed a countersuit against the Flints (later withdrawn, but initially based on the assertions that, on the day of his death, Kim Jr. had ­violated local traffic laws and, thus, the terms of his user agreement) and a motion for summary judgment—a request to have the case dismissed without a trial.

O’Melveny’s attorneys based the argument for dismissal on four principles ­supported by legal precedent: that cycling contains an implied assumption of risk; that, when he joined, Kim Flint expressly assumed these risks by agreeing to Strava’s terms and conditions (which included the words: “in no event shall [Strava] be liable to you or any third party for any direct, indirect, punitive, incidental, special or consequential damages arising out of or in any way connected with… your use or misuse of the site”); that Strava had no control over Flint’s activities, nor the situations and circumstances in which he rode, nor any knowledge of them before he uploaded data to the site; and that Strava was immune from liability under the Communications Decency Act (CDA).

The last refers to a landmark piece of federal legislation passed in 1996, originally to regulate pornography on the Internet but later amended to shield websites from suits that seek to hold them liable for content originating with “third parties”—i.e., users. According to Section 230 of this law, an interactive computer service cannot be treated as the publisher or speaker of information provided by an independent ­“content provider,” even when the content proves harmful offline. But the Flints’ attorneys argued that, unlike eBay or Amazon, content uploaded by Strava users doesn’t contain opinions or evaluations.

“There is nothing dangerous in and of itself about GPS coordinates and data,” said the family’s lawyer, Susan Kang. “The danger and harm alleged in this case originates out of Strava’s own actions in…manipulating it through its designed software into leaderboards, and then using those leaderboards to encourage cyclists to race at increasingly faster speeds for awards and titles.”

Moreover, Kang argued, Strava’s terms-of-service agreement constituted an ­unenforceable “adhesion contract,” as potential users “had no power to bargain the Terms with Strava and, since it was on the Internet, had literally no one to bargain with.” If affirmed, this would seem to invalidate every user contract for which any prospective website member has ever clicked “Agree.”

SUPERIOR COURT JUDGE Marla J. Miller nodded understandingly throughout Kang’s courtroom presentation. Then, as soon as the lawyers had finished talking, she ruled in favor of Strava, dismissing the case on the grounds that bicycling implies an assumed risk. As Michael Rustad explains, in contemporary American jurisprudence, such a decision represents the order of the day.

The Thomas F. Lambert Jr. Professor of Law at Boston’s Suffolk University, Rustad is currently completing a book on global Internet law. “Websites have no liability,” he says. “If there’s no possible legal sanction, even when use of a website results in death or injury, there’s no incentive to be careful.” Even if the case had gone to trial, Rustad says, the verdict probably would have favored Strava. “Assumption of risk is legally­ operative if the consumer understands a known risk [as Flint did]. Ski lodges require skiers to assume inherent risks, but no court would enforce a waiver if, say, the resort concealed concrete barriers or barbed wire close to the runs.” Moreover, he says, it’s a myth that juries are outrageously sympathetic to plaintiffs: “In reality, they’re disinclined to make awards, because there’s a blame-the-victim mentality in America. People think chowderheads get what they deserve—and that bicyclists are crazy anyway.”

Obviously, no one has to do anything just because a social network encourages them to. For reasons that had nothing to do with Strava, Kim Flint and Chris Bucchere enjoyed risking their lives on bikes—and when doing so were able to ignore the peril in which they put other people. Still, it’s natural to question whether the cyclists would have been as heedless if a website wasn’t posting leaderboards that tacitly rewarded their downhill daredevilry. “The more I think about this, it’s like Strava is creating a drag race,” Rustad says. “[Strava is] not just posting what third parties do—they’re organizing it. It goes beyond a passive display. A third party can’t do anything with the data without the website, so Strava takes on many attributes of a content creator. Its [undifferentiated-skill-level] leaderboards are comparable to taking people from the bunny slopes up to the black-diamond run. Even ski trails are marked by degrees of difficulty.”

IN AUGUST 2013, Chris Bucchere pled ­guilty to felony vehicular manslaughter in return for a sentence of three years’ probation, sparing him from prison but requiring him to perform 1,000 hours of community service. After six months, a judge can reduce Bucchere’s crime to a misdemeanor if he complies with the terms of his sentence.

Strava played no role in the proceedings, so neither Bucchere’s case nor the decision in Flint v. Strava Inc. addressed the bigger issues overhanging both: the power of the Internet and the dangers of unregulated online competition. Cycling (which takes place on machines, after all) has always been tied up with technology, but as the cyberworld continues to evolve at breakneck speed, few can foresee how it will go on to affect not just this sport but every aspect of our existence—enhancing, invading, and sometimes ending our lives.

In that murky atmosphere, Patrick Gordis watched with dread as the virtual and actual trajectories of Kim Flint—“a test case for Strava,” in the eyes of his friend Neal Trembath—merged in tragedy. “At some point,” Gordis reflects, “you lose sight of the fact that on Strava you’re not a video-game avatar who can endlessly respawn.”