Carnegie Mellon Solves 12-Year-Old DARPA Grand Challenge Mystery

Carnegie Mellon Solves 12-Year-Old DARPA Grand Challenge Mystery

Carnegie Mellon’s Red Team went into the 2005 DARPA Grand Challenge as the favorite to win. They’d led the pack in the 2004 event, and had been successfully running their two heavily modified autonomous Humvees, H1ghlander and Sandstorm, on mock races across the desert for weeks without any problems. When H1ghlander set out on the 212 km (132 mi) off-road course at dawn on 8 October 2005, it led the pack, and gradually pulled away from Stanford’s robot, Stanley.

About two hours into the race, however, H1ghlander’s engine began to falter, causing it to struggle in climbs and never reach its top speed. Nobody could tell what the issue was, but it slowed the vehicle down enough to cost it more than 40 minutes of race time. Stanley passed H1ghlander and went on to win the race by just 11 minutes. Even after the event, CMU wasn’t able to figure out exactly what happened. But last weekend, at an event celebrating the 10th anniversary of the DARPA Urban Challenge (which CMU won handily with their autonomous Chevy Tahoe BOSS), they accidentally stumbled onto what went wrong.

Here’s the point in the race where H1ghlander started to falter; it’s part of a fantastic NOVA documentary on the DARPA Grand Challenge which you should watch in its entirety if you have time:

Even the DARPA Grand Challenge winner, Stanford University, seemed a bit surprised by how things turned out. “It was a complete act of randomness that Stanley actually won,” Stanford team lead Sebastian Thrun later said. “It was really a failure of Carnegie Mellon’s engine that made us win, no more and no less than that.”

Here are some excerpts from Red Team’s race logs recorded immediately following the Grand Challenge:

October 11: The root cause that capped H1ghlander’s speed and crippled its climbs is not yet known. Requested speeds above 20 mph were under-achieved, even on the long, straight, level roads. H1ghlander didn’t even reach intended speeds going downhill. H1ghlander apparently stopped, rolled backwards, then re-climbed a few times. Weak climbing and stopping are not great practices for winning races. The capped speeds and weak climbs cost H1ghlander over 40 minutes of schedule time. The root cause is still a mystery.

October 12: H1ghlander’s engine was observed to be shaky immediately following the race. The first indication of possible engine trouble was observed when driving H1ghlander from the finish line to the inspection area with a human at the wheel. The engine was running very rough and almost died repeatedly in just that 50 yards of driving with a human foot on the accelerator pedal.

This engine problem is unlike any one that we have seen in the past, as engine performance is severely degraded at and anywhere near idle. Data indicated no limp home mode, no safety mode, and no low-torque mode. Detailed fuel, oil and transmission samples will be analyzed. We do not yet know the root cause that slowed H1ghlander’s driving on race day.

It turned out that the fuel was okay. The oil and transmission fluid were also okay. The electrical system was fine too. With the DARPA Urban Challenge up next and a completely new vehicle under development for that, the CMU team moved on.

Last week, CMU celebrated the 10th anniversary of BOSS’ DARPA Urban Challenge win in 2007. BOSS, Sandstorm, and H1ghlander were all pulled out of storage at CMU and tidied up a bit to be put on display. As H1ghlander’s engine compartment was being cleaned with the engine running, Spencer Spiker (CMU’s operations team leader for the DARPA challenges) leaned against the engine with his knee, and it started to die. This little box is what he was leaning against, as shown to Clint Kelly (who directed DARPA’s research programs in robotics and autonomous systems in the 1980s) by CMU Red Team leader William “Red” Whittaker.

The box is a filter that goes in between the engine control module and the fuel injectors, one of only two electronic parts in the engine on a 1986 Hummer. Spencer discovered that just touching the filter would cause the engine to lose power, and if you actually pushed on it, the engine died completely. But, from a cold start, if the filter wasn’t being touched, the engine would run fine. There was nothing wrong with H1ghlander’s sensors, or software: this filter cost H1ghlander 40 minutes of race time, and the win. “How about that, buddy!” Red said to Chris Urmson (who was working on perception for Red Team during the DARPA challenge, and ran Google’s self driving car program for seven years before starting his own autonomous vehicle company) at the CMU event, showing him the filter. “You’re off the hook!”

As to what may have caused this hardware failure in the first place, many team members at the CMU event suggested that it may have happened just a few weeks before the Grand Challenge, on September 19, when H1ghlander got into a bit of an accident after a 140-mile autonomous test:

Here’s an excerpt from a blog post by Vanessa Hodge, who worked on vehicle navigation and was following H1ghlander in a chase car that night:

H1ghlander was driving autonomously back to the entrance road so we could drive it back to the shop to pamper it before the race. We came to a part of the trail where there was a swamp on the left and a boulder-ridden mountain side on the right, with a road width a little bit larger than the vehicle. H1ghlander kicked up some thick dust and I slowed down to a stop to let the dust settle before catching up. The team in the chase car watched our vehicle display which monitors its actions while the dust settled. Problems appeared in the display, and one team member immediately hit the emergency pause button, but it was too late. In the second we lost visual, H1ghlander tracked off to the right of the path up the slope, slid on its side and flipped entirely to the other side.

By September 22, H1ghlander was back up and running, “hot and strong.” But, that filter may have taken some damage that was difficult or impossible to diagnose, and it ended up failing at the worst possible time.

While it’s impossible to know how a DARPA Grand Challenge win by H1ghlander might have changed autonomous car history, the people I spoke with at CMU generally seemed to feel like everything worked out for the best. BOSS had a convincing win at the DARPA Urban Challenge in 2007, and Stanley’s performance at the Grand Challenge helped to solidify Stanford’s place in the field. Roboticists from both CMU and Stanford helped to form the core of Google’s self-driving car program in California, and today, Pittsburgh is one of the places where both established companies and startups come to do self-driving research, development, and testing. There are very few lingering feelings about what happened, and everyone involved has long since moved on to bigger and better things. But all the same, it’s nice that at last, this final mystery has been solved.


How Cities Are Trying to Convince Landlords to Rent to the Homeless

How Cities Are Trying to Convince Landlords to Rent to the Homeless

Families wait years to get off the government’s waiting list for a rental voucher, sometimes while living in a homeless shelter. When they finally get that housing aid, they often struggle to find landlords willing to rent to them.

Most landlords screen out people who have a criminal background, poor credit or a history of evictions, making it difficult for voucher holders to find somewhere to live, even when they can afford rent. In fact, it’s common for people to lose their vouchers — which have expiration dates — after months of unsuccessful searching for a home.

To ease landlords’ worries and house more of the homeless, a growing number of cities are offering to reimburse landlords for certain losses — unpaid back rent or repairs for tenant-caused damages — that result from accepting applicants who have rental vouchers.

“Many, many communities are doing this, and it’s out of necessity,” says Elisha Harig-Blaine, who works on affordable housing issues at the National League of Cities. “They simply can’t get people placed into housing with these subsidies.”

This month, Boston and the District of Columbia announced their own “housing guarantee” or “risk mitigation” programs.

In Boston, the city will reimburse landlords for up to $10,000 in unpaid back rent or property damages that go beyond normal wear and tear. In D.C., a nonprofit is raising $500,000 in private funds to cover up to $5,000 in landlord costs per tenant. In both places, program staff will be available to address landlord complaints and provide case management for the tenants.

The question is, will that be enough to convince landlords to accept tenants who pay with rental vouchers?

In many of the cities that have these programs, affordable housing is hard to find, but renters with clean criminal and financial backgrounds are not.

“At the end of the day, real estate is a business. These landlords want to do the right thing, but we’re talking about their livelihood,” says Harig-Blaine, who has attended landlord recruitment events in nine communities across the country.

Landlords, he says, don’t want to deal with missed payments or other trouble that might come with renting to someone who was recently homeless.

Nevertheless, local officials in D.C. — which is getting 800 new residents every month and has some of the country’s highest rents — are optimistic.

“Rather than [renting] to the millennial who is just moving in from some other part of the country,” Neil Albert, president and executive director of the DowntownDC Business Improvement District, the nonprofit raising the money, told Governing. Albert thinks the risk funds will spur landlords to “weigh our needs and give equal consideration” to voucher holders.

There is no official number of landlord assistance programs, according to the U.S. Interagency Council on Homelessness, but they exist in Denver, Fargo, N.D.; Marin County, Calif.; Orlando, Fla.; Portland, Ore.; and Seattle-King County, Wash, which started one of the first almost a decade ago. Some states, such as Minnesota and Oregon, offer them as well.

Before launching its program, Boston researched them in other cities and found that participating landlords rarely had to use the risk funds, according to Boston’s Department of Neighborhood Development. Last year, in Seattle and King County, for example, participating landlords filed mitigation claims for only 15 percent of the renters covered by the program. Data on how many landlords participate in each city and how many people are housed through such programs, however, is not readily available.

D.C. officials, though, expect demand for the risk funds to be higher in their city.

“We think it will be a little different here in D.C. We think people will actually use this fund,” says Albert, adding that if it results in more units being rented to voucher-holders, then “that’s a great problem to have.”

One difference between the Boston and D.C. landlord programs is the funding and management structure. In Boston, the city is putting up the risk funds and managing its landlord relations on a two-year pilot basis. In D.C., a nonprofit business improvement district is raising funds — mostly from developers — and a local housing nonprofit is administering the program. That’s because landlords and property managers in D.C. pushed for a privately managed fund that could provide reimbursements faster than a government agency, says Albert.

More than 5 million Americans receive some kind of rental voucher from the state or federal government, according to the Center on Budget and Policy Priorities. To qualify, a person or household must be below the federal poverty line or make less than 30 percent of the area median income. Because the program is not an entitlement, less than a quarter of all eligible families receive housing assistance, and many households wait years before a voucher becomes available.

Landlord assistance programs are trying to address a chicken-and-egg problem, says Laura Zeilinger, D.C.’s director of human services. Landlords want renters who have jobs and earn a steady income. But stable housing is usually the first step to helping people get and keep a job. She’s hoping that landlords in D.C. will waive income requirements in their applications.

“Housing is an important foundation for people to be able to work and to achieve their potential,” she says. “It’s a really difficult thing for people to do while living in a shelter environment.”

The Dark History Behind Ouija Board’s Baltimore Origins

The Dark History Behind Ouija Board’s Baltimore Origins

Charles Kennard always had his eye out for a chance to make a buck, but he was not the greatest, nor the luckiest, businessman. It appears that he wasn’t the most honest guy, either. The second child of a successful Delaware merchant, Kennard moved to Maryland’s Eastern Shore in the late 1880s after developing “secret” bone-mix recipes for fertilizer. (In fairness, everyone in the fertilizer business claimed a “secret” recipe.) Following initial success, his Chestertown plant went to auction due to a combination of drought, competition, and debt. But all was not lost. A Prussian immigrant named E.C. Reiche kept an office next to Kennard’s on the first floor of the four-story, wood-frame hotel in Chestertown’s tiny business district. A furniture maker turned coffin maker turned undertaker—not an atypical career progression for the day—Reiche was also an inveterate tinkerer and Kennard had another plan.

Back story: Two generations earlier, a pair of girls in upstate New York named the Fox sisters, claiming to be mediums able to interpret mysterious “knocks” from the other side, had launched a spiritualist movement that continued to hold sway across the country. In fact, in the aftermath of the Civil War, with so many husbands, fathers, and sons lost in the conflict’s bloody battles, spiritualism—the belief the dead can speak to the living—had only gained steam with people desperate for a connection to departed loved ones and greater meaning for their own lives.

It’s in this context in 1886, during the period Kennard and Reiche shared a hallway, that newspaper reports began appearing about a “talking board” phenomenon sweeping Ohio, including an Associated Press story that ran in the local Kent County News. It’s also about this time, according a later Baltimore American story, that Kennard and Reiche—most likely inspired by the AP account—began collaborating and making at least a dozen of their own “talking” boards.

“Reiche, the biggest coffin maker in town, is making these on the side,” explains Robert Murch, the world’s foremost talking-board historian, and it’s these prototypes that became the Ouija board. “But it’s Kennard, when he leaves Chestertown for Baltimore in 1890, where he continues in the fertilizer game, and starts a real-estate business, who begins pitching what he says is his talking-board invention to potential investors.”

After numerous rejections, Elijah Bond, a local attorney who claimed his sister-in-law was a strong medium, finally took an interest. Soon enough, the Kennard Novelty Company, which incorporated the day before Halloween 125 years ago, began manufacturing Ouija boards much as they appear today. Bond was right about his sister-in-law, too: Helen Peters proved convincing enough with Kennard’s new talking board to win over a skeptical U.S. patent office. She not only gets credit for earning the stamp of legitimacy from the federal government, certifying the board delivered as promised, but also for “receiving” the O-U-I-J-A name from the board itself, which told her the strange word meant “good luck.”

(In truth, the name “Ouija” was written on the necklace locket that Peters was wearing at the time.)

So, yes, an undertaker and an opportunist named Kennard invented the only patented board game—billed as both a mystical oracle for communicating with the spirits and wholesome amusement—ever to outsell Monopoly in a given year.

“It comes straight from the 19th-century séances,” says Nic Ricketts, curator at The National Museum of Play in Rochester, NY, noting that a glow-in-the-dark board and a classic version are still sold today. “There has never been another brand board game like it, and I don’t see it fading away any time soon.”

The story of the Ouija board, however, is more than a tale of snake oil salesmen duping the Victorian masses or, subsequently, a game of harmless fun at a million junior-high sleepovers. While it remains an amazingly enduring pop-culture phenomenon—tied to the rise of the horror movie/paranormal industrial complex—its saga is also about the universal desire to find answers to life’s biggest questions, the history of psychology, and even the development of neuroscience.

“It’s always been a board game, a parlor game, but it has always been more than a board game for some people, too,” Murch says. “In the 19th century, people had a much different relationship to death than we do today—it was much closer to their everyday experience. Now, we do everything we can in hopes of avoiding aging, let alone engage in any real thoughts of death. But in the 1800s, people only lived to be 50 years old. Mothers would have 12 children and six of them would die. Their parlor rooms were also their funeral rooms.”

Not surprisingly perhaps, there’s a dark side or two buried in Ouija’s origin story. There always is when money is at stake, and by the early 1890s, some 2,000 Ouija boards were already being sold a week. William Fuld, who worked for and invested in the Kennard Novelty Company—and eventually gained control of the Ouija business after the founder cashed out too early—went on to make millions manufacturing the board in Baltimore and elsewhere, but only after his brother was cut out of the company. Their ensuing lawsuits were no mere spat. William’s brother, Isaac, became so embittered that he had his baby daughter exhumed and relocated from the Fuld family gravesite during a cemetery renovation. The two sides of the family would not speak for 96 years.

And, tragically, William Fuld would suffer a fatal accident at his Harford Avenue factory, one he claimed in a 1919 Baltimore Sun story that the Ouija had told him to build. (“Prepare for big business.”) Overseeing the installation of a flag, an iron railing gave way and he fell off the roof of the structure, which still stands and has been converted into a senior apartment complex. “On his death bed—the coroner’s report said a broken rib pierced his heart—he made his children promise to never sell the Ouija out the family,” says Murch.

Of course, Fuld’s family did sell—but not for four decades—to Parker Brothers, which promptly moved Ouija to its base of operations in Salem, MA. In 1967, the first year it was headquartered in the town infamous for its witch trials, Ouija sold two million boards.

By comparison, Monopoly—an early version was invented in 1903—wasn’t popular until the Great Depression, when it fulfilled a kind of fantasy escapism. Ouija, on the other hand, was a sensation from the outset, long before even its first film appearances, which date back to Hollywood’s beginnings.

But Ouija’s public image has always been complicated. Initially, the “mysterious oracle” was marketed as a game to enliven a party or encourage a little light-hearted intimacy for romantic—or would-be romantic—couples, who are often depicted in early advertisements with the board resting on their knees as they sit across from each other, both of their hands on the planchette. Norman Rockwell, who was fond of depicting the revealing moments of everyday life, painted a well-dressed suitor and young woman, chairs pulled face-to-face, playing with a Ouija board for the cover of The Saturday Evening Post in 1920.

Less well known is the Ouija board’s use as inspiration or as an “automatic” writing tool by acclaimed novelists and poets, such as Sylvia Plath, who wrote “Dialogue over a Ouija Board,” and Pulitzer Prize winner James Merrill. Merrill used notes from Ouija “consultations” in his 560-page epic poem, The Changing Light at Sandover, which contained messages from W.B. Yeats, friend Maya Deren, and the Archangel Michael.

But over time, the relative innocence of the Ouija board—or at least its nonpartisan relationship between good and evil—gave way to a more sinister reputation as Hollywood began utilizing it for darker purposes. After The Exorcist, in which actress Linda Blair’s character Regan explains to her mom, played by Ellen Burstyn, how she used the family’s Ouija board to ask questions of “Captain Howdy”—the demon who eventually takes possession of her soul—the board’s occult status was cemented.

Since then, it has shown up in more than 20 films, and made countless appearances in the ever-growing number of paranormal-themed TV shows. Forums around Ouija-associated phenomena populate the Internet, of course. Most recently, the 2014 movie Ouija did so well at the box office that Ouija 2 is already in the works. When it was released last fall, the movie so dramatically boosted board sales that petitions by evangelical Christian groups to ban the Ouija started popping up again., a lay-run Catholic apologetics and evangelization website, describes Ouija as “far from harmless.”

Still, the most interesting thing about the Ouija board might be the latest research around it from University of British Columbia that shows it actually does work—just not in the way we might assume.

A few years ago, Sidney Fels, professor of electrical and computer engineering at UBC, brought out a Ouija board at a Halloween party attended by graduate students, including many who were foreign-born and unfamiliar with how it works. They assumed it required batteries. “‘No, you don’t need batteries. It will move,’ I told them,” Fels recalls. “I gave them some mystical explanation tied into Halloween and they had a good laugh.”

But lo and behold, when Fels returned later, the grad students were enthralled because the planchette was moving on its own. Or so it appeared. The mechanism at work was actually something known as the ideomotor effect, which refers to the influence of the unconscious mind on muscle movements. (First identified in 1852, preceding Sigmund Freud’s theory of the unconscious mind by decades, Dr. William Benjamin Carpenter discovered the ideomotor effect while investigating the unconscious mind’s ability to direct motor activity. Shortly thereafter, other researchers began linking that discovery to—you guessed it—spiritual phenomena.)

Days later, still fascinated by the students’ experience, Fels shared the story with colleague Ron Rensink, a psychology and computer science professor, and that got the ball rolling about whether the board could serve as a tool to look at unconscious knowledge.

“We didn’t know if we’d find anything, but when we did, the results really surprised us,” Fels says. When study participants were asked to answer or guess at a set of challenging questions, they were correct about 50 percent of the time. But when responding while using the board—which participants believed had the ability to “receive” correct answers from another person teleconferencing via a robot Ouija partner—they scored correctly upwards of 65 percent of the time.

In actuality, the robot was a ruse; it was not responding to the video-conferencing player, but subtly amplifying the study participants’ tiny, unconscious movements. “It was significant how much better they did on these questions,” Rensink says. “If you don’t think so, consider the difference playing roulette when the odds are 50-50 versus 65-35.”

The implication is that one’s unconscious is much smarter than anyone knew, capable of pulling up bits of stored information not accessible to the conscious mind.

Results in a follow-up study replicated the findings, which they reported in the academic journal, Consciousness and Cognition.

Rensink believes the results open greater possibilities for further study. For example, is unconscious memory affected by Alzheimer’s and other neurodegenerative diseases in the same way as conscious memory?

It’s work that William Fuld—the guy who fell from the factory roof and is considered the “father” of the Ouija (he was also a state delegate and philanthropist)—would probably appreciate. When asked directly by a reporter if he believed in the Ouija’s mystical powers, he replied: “I should say not. I’m no spiritualist. I’m a Presbyterian.”

The discovery of the Ouija’s ability to tap into unconscious knowledge is not the only development in the talking board’s 125-year-old story, however. The reconciliation of William Fuld’s family with his brother Isaac’s clan after nearly a century of silence is the other compelling occurrence.

The two sides had long lost contact until Murch began posting his research on the web nearly two decades ago. That’s when Stuart Fuld, the then-sixtysomething grandson of Isaac Fuld, and Kathy Fuld, the granddaughter of William Fuld, separately reached out to Murch, in hopes of learning more about their ancestors. “I was talking to each one individually at first without the other one knowing it,” Murch recalls. “I was aware of the feud and didn’t want to upset either one, but then Kathy called one night and asked for Stuart’s phone number.”

“It turned out we were living five miles apart while growing up and didn’t know it,” says Kathleen Fuld, Stuart’s wife. (Stuart Fuld passed last year.)

The two sides of the family, which now include great-grandchildren and great-great grandchildren of the brothers, have been getting together regularly ever since.

While some of the descendants did hold on to Ouija and other talking-board memorabilia—Isaac later attempted to launch a talking board competitor named the “Oriole” board—no one, apparently, ever took a serious spiritual interest in Ouija. Not even when they were kids.

“Not me,” says Kathleen Fuld, chuckling. “I was a good Irish Catholic girl.
I had eight cousins who were nuns.”

She adds, however, that Stuart did take a great deal of interest in learning about his grandfather and ancestors, as well as the history of the former family business—if not the surrounding mysticism—especially as he got older.

“I’ll tell you a funny story,” she says. “We went up to the Poconos for a golfing trip one year and there was a conference of priests taking place at the hotel where we stayed. I don’t remember why or how it came up, but Stuart ends up telling
a group of priests we’re talking with that his family once made the Ouija board.

“All the priests immediately started making little crosses with their fingers,” Fuld continues. “They started asking Stuart all kinds of questions. They wanted to know the whole story and got the biggest kick out of that.”

Even better, the priests invited the couple to take advantage of the conference’s complimentary evening cocktail parties for the weekend—which they did.

“But it didn’t matter,” she adds. “Every time we saw those priests, in the elevator, or wherever, they’d start making those crosses with their fingers.”

These Two Small Letters Heralded the Beginning of Online Communication

These Two Small Letters Heralded the Beginning of Online Communication

An uncountable number of letters have been sent from one person to the other via the internet in the years since 1969–in ARPANET message boards, the recently-deceased AOL Instant Messenger and currently-in-vogue Slack, to name a few platforms. Hard to believe, but this communication revolution started with two letters.

Late at night on October 29, 1969, today celebrated as International Internet Day, the first message was sent over the Internet. Two groups of researchers in two separate facilities sat before rudimentary computer terminals, on the phone, making yet another attempt at talking to each other. Their planned first transmission wasn’t anything too fancy, Len Kleinrock, who headed the UCLA lab engaged in the research, told Guy Raz for NPR. But it turned out to be amazing anyways.

The UCLA researchers were trying to transmit the message “login,” as in a login command, to the computer at Stanford. Charley Kline, who sent the initial transmission from UCLA, said they’d tried this before with no success. This time, however, something happened. “The first thing I typed was an L,” he told NPR. Stanford computer scientist Bill Duvall said over the phone that he’d received it. He typed the O: it also went through. Then came the G: “And then he had a bug and it crashed.”

Later that night, after some more tinkering, they successfully transmitted the whole word. Then they went home to get some sleep, having no way of knowing what would ensue because of this development.

“We should have prepared a wonderful message,” Kleinrock told Raz. It would have placed them in the tradition of discoverers who had pithy statements– “What hath God wrought,” “a giant leap for mankind,” etcetera. Samuel Morse, Neil Armstrong and the others “were smart. They understood public relations. They had quotes ready for history.”

But “lo,” the accidentally abbreviated first transmission, would have to do, and in fact actually works quite well. Merriam-Webster defines the word as an exclamation “used to call attention or to express wonder or surprise” that has a history of use going back as far as the 12th century. Its predecessor, the Middle English “la,” goes back even farther. According to the Oxford English Dictionary, “la” can be found in Beowulf and the Ormulum, amongst other works. Its more modern incarnation is found in the King James Bible, in the first scene of Hamlet and in Tennessee Williams’s A Streetcar Named Desire, to name a few examples.

What the teams at UCLA and Stanford had pioneered was the ARPANET, the predecessor to the internet, which has come to contain all of the above texts as well as many, many more pedestrian statements. By the spring of 1971, it could be found at 19 research institutions, writes Leo Beranek for the Massachusetts Historical Review, and it’s only spread from there.

Two identical nuclear projects, one in Georgia and one in South Carolina. Only one survived.

WAYNESBORO, GA. – At 4:30 p.m., a siren sounds at the Vogtle nuclear worksite, and the exodus is on.

Workers from across the nation have moved here to build two nuclear reactors, so from Missouri and Texas, welders and electricians pile into cars and trucks. From Georgia and South Carolina, they form a line out the gates to a country road that’s stop and go.

Georgia’s Vogtle nuclear reactor project is just 130 miles from South Carolina’s V.C. Summer Nuclear Station, but activity at the two complexes couldn’t be more different.

At South Carolina’s V.C. Summer site north of Columbia, parking lots are mostly empty. Many areas resemble junkyards after workers there lost their jobs and walked off the site in July. Left behind: trucks with their doors wide open; construction plans scattering in the wind, and tens of millions of dollars in equipment and supplies, now rusting and rotting away.

But here in the pine lands and cotton fields south of Augusta, nearly 6,000 workers still have jobs. And, at shift change on this rainy Monday afternoon, they create an hour-long traffic jam. Behind them, a dozen cranes rise from a vast parking lot that’s still packed. Two new cooling towers stand so high they merge with rain clouds.

The Vogtle and V.C. Summer nuclear projects were identical twins. Both had the same reactor design, the same contractor, the same modular construction strategy. They began at roughly the same time in the late 2000s. Both benefited from new laws that shifted financial risks from utilities to ratepayers. But only one of these twins survived. Why?

A deeper look at the Vogtle nuclear project helps explain what happened to its South Carolina twin, how Westinghouse bungled construction at both sites. It exposes South Carolina’s lax oversight and how this failed to give ratepayers a true picture of V.C. Summer’s problems. Georgia, meantime, had monitors who consistently and painstakingly pointed out the project’s problems — in public. Ultimately, Vogtle’s story shows how money and political power triumphed over incompetence, at least so far.

Nuclear renaissance

The Vogtle nuclear complex is roughly halfway between Allendale and Augusta, just off a two-lane road that roughly follows the Savannah River. Amid the area’s rolling hills are a scattering of modest homes, dozens of RV parks for workers and two convenience stores. A prodigious amount of roadkill is a testament to its remote location and heavy construction traffic.

Southern Company’s subsidiary, Georgia Power, began work here in 1974 on two nuclear reactors, Vogtle 1 and 2. The company named them after Alvin W. Vogtle, a former chairman and president, who among other things was a World War II prisoner of war. The feisty Vogtle tried to break out four times, succeeding on his fifth. His exploits inspired Steve McQueen’s character in the movie “The Great Escape.”

Georgia Power’s relationship with nuclear power isn’t so different than its persistent former chairman. In 1979, while Vogtle 1 and 2 units were still under construction, Pennsylvania’s Three Mile Island plant had a near meltdown. Public support for nuclear power cooled, and chilled even more after Russia’s Chernobyl blast in 1986. But Georgia Power kept pushing forward. Vogtle 1 finally went online in 1987, and Unit 2 in 1989.

By then, America’s nuclear construction industry had begun a two-decade hibernation. Burned by high costs and numerous failures, energy producers moved toward coal and natural gas.

But in the early 2000s, Georgia Power and other utilities were ready to try again. In 2003, a little-known group, NuStart Energy, formed to usher in a new era of nuclear power in the United States. The group included Southern Company, SCANA and nine other utilities, mostly in the South. Joining them were Westinghouse and General Electric, each with a new reactor design. They stood to make tens of billions of dollars if this nuclear renaissance materialized.

The power of Georgia Power

Things would be different this time, Westinghouse promised then and still does today. Westinghouse had a new reactor design, the AP1000. It incorporated the latest technology. Old reactors use pumps to shower water onto overheating reactors, but the AP1000 relied on a nearly 800,000-gallon container of water above the reactor. In an emergency, water would pour onto the reactor and prevent a meltdown.

And new plants would be constructed differently, Westinghouse advertised. Instead of doing nearly all work onsite, key components would be fabricated off site at dedicated factories. These modules would then be shipped to work sites and assembled like giant Legos. This approach would cut costs and improve quality, especially if Westinghouse produced reactors in large numbers as the industry hoped.

Utilities and their lobbyists went to work. In the mid 2000s, they prodded state legislatures to pass pro-nuclear laws. These new laws, including South Carolina’s Base Load Review Act, generally shifted financial risks of new reactors from utilities to ratepayers.

The industry’s campaign was particularly aggressive in Georgia. Energy-related PACs and lobbyists poured money into campaign coffers of lawmakers and utility regulators alike.

Unlike South Carolina, where lawmakers appoint commissioners, Georgia’s commissioners are elected. Campaign donations even crossed state lines, with SCANA executives chipping in at least $53,000 since 2006, according to a compilation by the Energy and Policy Institute, a watchdog group. All told, lobbyists and utility players pumped nearly $2 million into commissioners’ campaigns over the past decade.

Georgia Power “has deep tentacles and deep relationships” at all levels of government, said Glenn Richardson, Georgia’s House Speaker from 2004 to 2010, the first Republican since Reconstruction to hold the post. He said Georgia Power regularly showered communities with donations, accumulating goodwill and favors. It had a deep stable of lobbyists. Richardson said he was amazed at how the utility could advance its legislative interests quickly, often without leaving fingerprints.

Yet, he added, Georgia Power and its allies made solid cases that they needed new nuclear reactors. Coal was in decline; natural gas prices were still high in the mid-2000s; solar and wind weren’t yet competitive. “And, you have to remember that in 2004, ’05 and ’06, Georgia had exponential growth,” he said. “We were on fire, and we didn’t think we would have enough electricity for everyone moving here.”

Amid this legislative push, on April 9, 2008, a consortium led by Georgia Power made a deal with Westinghouse to build two new AP1000 reactors at Vogtle.

The next month, SCANA and Santee Cooper made a similar deal for V.C. Summer near the small town of Jenkinsville.

The race to build the new generation of nuclear reactors had begun.

The insiders

Constructing a plant to safely split atoms is a remarkably complex endeavor, one that requires extensive engineering know-how and documentation. So, in 2009, Georgia’s Public Service Commission set aside $600,000 a year to hire expert construction monitors. Georgia Power agreed to foot the bill. The commission then hired William Jacobs to lead the monitoring team.

Jacobs had a lengthy resume: He graduated from Georgia Tech with a Ph.D. in nuclear engineering. He worked for a Florida engineering company and on nuclear projects in Korea. In 1979, he landed a job with Westinghouse and managed nuclear plants in the Philippines and Yugoslavia. In the late 1980s, he became a sought-after consultant, offering testimony about nuclear issues in at least 10 states, including South Carolina.

South Carolina regulators tried to hire Jacobs to monitor the V.C. Summer project but lost out to Georgia, said Nanette Edwards, deputy executive director of South Carolina’s Office of Regulatory Staff.

On the surface, Georgia’s monitoring plan had potential conflicts of interest, given that Georgia Power was effectively paying the monitoring team’s expenses and Jacobs’ past employment with Westinghouse. But Jacobs knew that a nuclear renaissance wouldn’t happen unless Westinghouse built Vogtle on time and on budget.

So, as construction picked up in 2012, Jacobs and his team aggressively identified roadblocks — ones that would haunt both the Georgia and South Carolina projects in the coming years. In public hearings that year, Jacobs warned that Westinghouse had no firm schedule that integrated the many tasks needed to complete the reactors. Instead, Westinghouse managed the project with a “six-month lookahead schedule.”

It was a stunning revelation; the absence of a firm schedule was particularly critical given the project’s modular construction. Delays in one module could ripple through the entire project. And Jacobs highlighted major quality-control problems at off-site factories. He cited safety problems in Louisiana uncovered by the Nuclear Regulatory Commission. Welders and workers there didn’t have the expertise to build nuclear components, much less document what they were doing. He testified that basic paperwork lacked signatures and contained illegible notes. Documentation was in such disarray that sorting it out would take longer than building a module from scratch.

Work at the Vogtle site itself also wasn’t going well; early concrete pours weren’t done properly, he told commissioners. Workers had to redo rebar because it wasn’t done to code. During a 2012 hearing, he said that Westinghouse’s performance was unsatisfactory — so bad that he urged Georgia regulators to consider a surprising scenario: the project, though barely off the ground, already could be as much as four years behind schedule.

Different state, different monitor

In 2011, South Carolina’s Office of Regulatory Staff also hired a construction monitor, another nuclear industry insider named Gary Jones.

Like Jacobs, Jones had an impressive resume. A senior vice president and owner of Sargent & Lundy, a large Chicago construction consulting company, he worked on nuclear projects in China and Korea and several in the United States.

Unlike Jacobs, Jones painted a much rosier portrait when he testified about the project’s status in 2012.

He told the S.C. Public Service Commission that V.C. Summer project hadn’t produced a detailed project schedule, but that he expected one soon.

At the time, SCANA wanted to push the completion date seven months forward. Jones warned commissioners this timetable was “aggressive and ambitious” and noted problems with module fabrication and quality control issues. But he concluded with a recommendation that regulators accept South Carolina Electric & Gas’s aggressive new completion schedule — and that the state approve SCE&G’s request to increase the cost of the project by $131 million.

He wouldn’t testify publicly for another four years.

Both projects fall behind

By 2016, Jacobs and his colleagues had a long history of identifying and making public major problems at the Vogtle site.

Time and again, he and his colleagues found that Westinghouse consistently failed to produce a detailed construction schedule, depending instead on short-term “look-aheads.”

Over and over he brought up that Westinghouse had yet to finalize engineering drawings. He told Georgia commissioners about his visit to Westinghouse’s AP1000 reactor project in China. His biggest impression from that trip was how far workers at Vogtle had to go.

He testified roughly every six months. Information about his findings was posted on the Georgia Public Service Commission website, as well as the questions his team posed to Georgia Power. These public hearings also gave the project’s opponents chances to cross examine Georgia Power officials. Armed with Jacobs’ findings, a lawyer for the Southern Alliance for Clean Energy grilled a Georgia Power official earlier this year about the project’s slow pace.

The official acknowledged that a consultant had examined productivity and found workers spent more time doing “non-work-related activities” than actual work. Progress was so slow that between April 2016 and April 2017, the Vogtle project fell behind 325 days, or “almost a day-for-day slippage,” the Southern Alliance for Clean Energy noted in a court filing.

Vogtle’s twin in South Carolina had similar productivity issues and delays. But the V.C. Summer project’s problems were often papered over by SCANA and Santee Cooper officials. On Sept. 21, 2016, SCANA and Santee Cooper held a “media day” at the plant where executives assured reporters that the project had faced some challenges but was on track.

“People ask, ‘If you could do it again, would you make the same decision?’” SCANA Chief Executive Officer Kevin Marsh told reporters. “Absolutely, I would make the same decision. I feel as strongly today — probably even stronger today than I did back in 2008 — that this is the solution for us, for a clean energy future.”

Behind closed doors, however, SCANA and Santee Cooper officials had received a scathing report from a consultant, Bechtel, a large civil engineering company. Kept secret from investors and the public, the Bechtel audit highlighted many problems that Jacobs and his team in Georgia had talked about publicly for years.

‘Significant progress’

South Carolina’s Office of Regulatory Staff has paid Jones, its construction monitor, an average of $122,000 a year since 2012 to keep track of work at V.C. Summer. But Jones reportedly didn’t learn about the Bechtel report until this summer when it was referenced during a Senate hearing and posted on The Post and Courier’s website, said Edwards of the Office of Regulatory Staff.

Jones had visited the V.C. Summer site monthly since 2011 and provided the Office of Regulatory Staff regular reports, ORS officials told The Post and Courier. But he testified publicly just twice, once in 2012 and again in 2016.

“We feel very good about his testimony in 2016,” Edwards said.

Jones’ 2016 testimony wasn’t as rosy as his comments four years before, but it still lacked the detail and sense of urgency of Jacobs’ work in Georgia. Jones testified about the project’s “significant progress,” especially now that Westinghouse had brought in Fluor, an experienced construction company, to sort out the project’s longstanding problems. But he added that he was skeptical about Westinghouse’s ability to complete the project on time.

“Well, I’ll be perfectly candid,” Jones told state regulators in 2016. “I don’t think that the project is going to hit their specific completion dates.”

Jones did not respond directly to recent requests for comment, but in an email through the Office of Regulatory Staff, he said Westinghouse had a schedule that changed often because of delays. He said that SCE&G and Westinghouse couldn’t agree on a long-term labor and construction schedule in 2013, and that “the project was managed on the basis of detailed three-month look-ahead schedules” after that. As the months passed, ORS never gained access to a long-term schedule, he said.

By early March, the problems with South Carolina’s project couldn’t be hidden anymore. At a meeting in New York, Westinghouse officials told SCANA and Santee Cooper that it still didn’t have a detailed construction schedule.

Lonnie Carter, Santee Cooper’s CEO, later wrote a letter calling the news “devastating.” Devastating perhaps, but given Jacobs’ public testimony about Vogtle’s schedule issues, it shouldn’t have been surprising.

Spreading the pain

Stan Wise, chairman of Georgia Public Service Commission, watched with interest as SCANA and Santee Cooper wrestled with their decision to shut down the V.C. Summer project. SCANA and Santee Cooper and their ratepayers had sunk $9 billion into the South Carolina project, more than the state’s annual budget. Westinghouse had bought more than 90 percent of the supplies for the plant but was less than a third of the way through actual construction. Estimates about the final price tag topped $25 billion.

When SCANA and Santee Cooper finally pulled the plug July 31, Wise said to himself: “We’re the last one standing.”

Georgia Power’s consortium and Georgia regulators also had a decision to make. As with V.C. Summer, only about 30 percent of Vogtle had been built.

“There was some real gnashing of teeth in Georgia to go or not go,” Wise said. But in the end, the Georgia consortium replaced Westinghouse with Bechtel as its primary contractor and pressed on.

While the Vogtle and V.C. Summer projects were virtually identical, Wise said the utilities behind the projects were notably different, which partly explains why Georgia Power moved forward.

The Georgia nuclear consortium received more money from Toshiba to settle construction problems. Georgia also had loan guarantees from the federal government that cut future financing costs. SCANA and Santee Cooper had decided against such guarantees.

Moreover, Georgia Power and its parent, the Southern Company, was much larger than SCANA, and could spread the project’s cost over a larger base of ratepayers, Wise said.

Georgia Power serves 2.5 million customers versus SCANA’s 717,000. In South Carolina, roughly 18 cents per dollar on an SCE&G bill goes toward the nuclear project. Because of Georgia Power’s larger customer base, it’s just 5 cents per dollar for its customers.

Wise called the difference in ratepayer risks “remarkable.” The 18 percent charge on SCE&G electric bills “will have a terrible impact on South Carolina residents and businesses for decades to come,” he said.

“We’re better off today with it going forward,” he added, “In South Carolina you have a finger pointing campaign about who knew what and when and who’s going to pay for it all. No good comes out of any of those scenarios.”

Quitting time

Monday was the first day that Bechtel was officially in charge of construction at the Vogtle site.

“They can’t point fingers anymore at Westinghouse,” Wise said. “It’s theirs.”

It was an inauspicious beginning for Bechtel. The day’s rainstorm drenched the worksite, and a few departing workers said they were let off early. They left during the afternoon shift change. And the result was like a parking lot after a major sports event: a massive rush to get out.

With the new cooling towers getting soaked by the storm, the huge, transient workforce filed out of two gates. From Kentucky and Alabama, they turned onto River Road, a few stopping at convenience stores to pick up beers. From Florida and South Carolina, they sped on country roads, turning into the many RV parks that have popped up like toadstools around the plant. Thousands of workers on the move, the last ones standing.

The Radical Reference Librarians Who Use Info to Challenge Authority – Atlas Obscura

The Radical Reference Librarians Who Use Info to Challenge Authority – Atlas Obscura

An adaptation of Banksy’s “Flower Bomber,” this depicts a librarian in protest, throwing Margaret Atwood’s A Handmaid’s Tale.

FROM AUGUST 29 THROUGH SEPTEMBER 2, 2004, a series of protests erupted in New York in response to the 2004 Republican National Convention and the nomination of George W. Bush for the impending election. Nearly 1,800 protesters were arrested during the convention, and later filed a civil rights suit, citing violation of their constitutional rights.

During the protests, a steady team provided support to anyone who needed information amid the confusion: a modest group of socially conscious librarians from around the United States, armed with folders of facts ranging from legal rights in dealing with police to the locations of open bathrooms.

“We wanted to operate as if we were bringing a reference desk to the streets,” explains Lia Friedman, Director of Learning Services at University of California San Diego, who was at one of the protest marches in 2004. At the time, fewer people had smartphones, making this service both new and important. When someone asked a question that wasn’t included in their traveling reference desk folders, other librarians waiting at their home computers were poised to research and deliver information by phone.

The group of librarians soon formed into the first-ever chapter of the Radical Reference Collective, a non-hierarchal volunteer collective who believe in supporting social justice, independent journalists, and activist causes. Since the group’s first action at the Republican National Convention of 2004, the group, originally based in New York, has spread across the United States as a collection of individual local chapters. New collectives formed via library listservs, the Rad Ref’s website, and word of mouth.

Because the organization is non-hierarchical, there is no consensus on exactly what the “radical” in Radical Reference really means, nor what kind of work it might imply. Members come from different backgrounds; one New York City Rad Ref member is personally involved with a group that provides street medics training to help people medically during protests; other members are from academia or are involved in work with prisoners or science archives. The website for Rad Ref points out that in no way does the word “radical” specifically denote a political affiliation of any kind; the word “radical” is used to challenge “mainstream meaning which largely marginalizes the term and along with it certain groups.” But, members do form their own opinions on the matter.

“In my opinion, using the word ‘radical’ means advocating for change, whether that is political, societal,” says Friedman. Audrey Lorberfeld, Digital Technical Specialist at The New York Academy of Medicine and longtime member of the New York Rad Ref group, says that this inherent politicization of information is apparent to librarians regardless of their specialty: in the introductory information sciences courses for those pursuing a Masters of Library Sciences in the U.S., librarians-to-be become well-versed in the American Librarians Association code of ethics, which includes intellectual and informational freedom. According to both Lorberfeld and Friedman, who were interviewed separately for this article, to many librarians, the idea that information is neutral is a myth.

Now, amid a divisive political climate in the U.S., the original New York group is continuing to provide open-access information for all, be it about a specific historical fact, civil rights infringement statistic, or the complex laws regarding immigration. Often this support is lended to social justice organizations, independent journalists, and, as their website states, “anyone who questions authority.”

Some of the Rad Ref groups use social media to communicate and promote events, talks, and workshops with the public and activist organizations, while others meet face-to-face. The topics vary wildly; some events may concern more local issues, like planning support for a library worker strike, while others could involve “creating or using a resource guide on a relevant issue, i.e. Black Lives Matter, Critical Librarianship, Fact Checking basics” says Friedman.

Rad Ref has sizzled in the background of protests, local workshops and activist groups across the U.S. since its inception, although participation varies, and each local group is unique. Friedman spent years providing information through the Rad Ref website, which formerly acted as a virtual reference question desk. More recently she has participated on a smaller scale within the San Diego group, which sometimes has only a few members; both Friedman and Lorberfeld noted that since librarians tend to be involved in multiple projects and membership is voluntary, the numbers of a group can fluctuate. Occasionally, the groups have gone on hiatus, as the New York group’s online reference presence did in 2013, when its members didn’t have time to devote to the struggles of running a volunteer-based organization.

That hiatus hasn’t lasted, though: New York City’s Rad Ref was reinvigorated after President Donald Trump took his oath of office in early 2017. And these librarians are ready to radicalize their role as information champions. During their first meeting post-hiatus, the room was overflowing with activists and librarians who deeply cared about organizing to preserve information. “It was an amazing feeling,” says Lorberfeld.

Right now, the New York group’s goal is to build a community of knowledgeable experts on the art of finding and delivering information, which can become a resource for librarians and activist organizations alike. “We’re trying to look inward and educate ourselves, from anything from immigration rights to grassroots organizing best practices,” says Lorberfeld. “One member is making a guide about Unions in New York City; what they are, how to join them,” and another is creating a map of available free meeting places for organizations throughout the city.

Friedman and other Rad Ref volunteers once helped a writer who was working on research for a nonfiction book on resistance and struggle within women’s prisons, for example, and needed access to statistics and facts that were not easy to find, such as the amount of prisoners in a specific county in a specific year. Through the website, Rad Ref librarians were able to provide specific numbers she used throughout her book by pooling their skills in data research and law. While the reference service aspect of the website has been inactive since the hiatus, the website is still available as an archive, and librarians will still answer the occasional question in the Rad Ref inbox.

It might seem like information is open to everyone today: the internet is common and most people know how to type a sentence into Google. But as Friedman and Lorberfeld explain, there’s often more to it than that when trying to find specific and often very meaningful details and information. Sometimes, the only thing keeping data from, say, a government website like that of the Environmental Protection Agency from being removed or tampered with by politicians may be the librarians, working behind the scenes to preserve that information—something that actually happened earlier this year.

“We used to teach that a .gov site was trusted. And that’s a little bit more challenging to do now,” Friedman says. A Google search often isn’t curated or necessarily fact-checked, and doesn’t always provide multiple and balanced sources. Sometimes key information is behind an academic journal’s steep paywall, or buried in government documents under specialized lingo. “We really wanted to support independent journalists and activists, and really wanted to give people access to information, which is a pretty librarian thing to do,” says Friedman.

It really is a pretty librarian thing to do. Despite librarians’ public image of glasses-clad women hushing pesky kids, library workers around the world, from South Africa to Sweden, have formed similar organizations to Rad Ref, according to Alfred Kagan in his book Progressive Library Organizations: A Worldwide History. Since the 1960s in the United States, many library workers have committed to “social responsibility,” the democratic giving of information to the public and to free speech. Kagan writes that progressive groups have used their independence from the American Librarians Association (ALA), the major national librarian group in the U.S., “to take radical stances.”

A subunit of the ALA includes the Social Responsibilities Round Table, which has, according to its website, worked to democratize the ALA with human and economic rights in mind since 1969. Similar groups exist around the United States, such as the Progressive Librarians Guild, which aims toward an international agenda. Rad Ref participates in this tradition, though it differs in that it’s comprised of local librarian groups who work within their individual volunteer base’s skills and goals, without a central governance.

“We all sort of have a really core sentimental belief of information access as a human right, and I feel like that really governs what we do within the group,” Lorberfeld says. By using their diverse backgrounds and talents, Rad Ref is now readying itself to help activist groups by promoting information that aids in furthering social equality.

Algorithms Have Already Gone Rogue | WIRED

Algorithms Have Already Gone Rogue | WIRED

For more than two decades, Tim O’Reilly has been the conscience of the tech industry. Originally a publisher of technical manuals, he was among the first to perceive both the societal and commercial value of the internet—and as he transformed his business, he drew upon his education in the classics to apply a moral yardstick to what was happening in tech. He has been a champion of open-source, open-government, and, well, just about everything else that begins with “open.”

His new book WTF: What’s the Future and Why It’s Up to Us seizes on this singular moment in history, in which just about everything makes us say “WTF?”, invoking a word that isn’t “future.” Ever the optimist, O’Reilly celebrates technology’s ability to create magic—but he doesn’t shirk from its dangerous consequences. I got to know Tim when writing a profile of him in 2005, and have never been bored by a conversation. This one touches on the effects of Uber’s behavior and misbehavior, why capitalism is like a rogue AI, and whether Jeff Bezos might be worth voting for in the next election.

Steven Levy: Your book appears at a time when many people who once had good feelings towards technology are now questioning it. Would you defend it?

Tim O’Reilly: I like the title WTF because it can be an expression of amazement and delight or an expression of amazement and dismay. Tech is bringing us both. It has enhanced productivity and made us all richer. I don’t think I would like to roll back the clock.
Not that rolling it back is an option.

No, but it’s important for us to realize that technology is not just about efficiency. It’s about taking these new capabilities that we have and doing more with them. When you do that, you actually increase employment. As people came off the farm, we didn’t end up with a vast leisure class while two percent of people were feeding slop to animals. We ended up creating new kinds of employment, and we used that productivity actually to enhance the quality and the quantity of food. Why should it be different in this era of cognitive enhancement? Uber and Lyft are telling us that things we used to think of as being in the purely digital realm, in the realm of media, whatever, are coming to the real world. So that’s the first wake up call for society. Secondly, we’re seeing a new kind of interaction between people and algorithmic systems. Third, they represent a new kind of marketplaces based on platforms [in this case, they exist because the of the platform of smartphones—and then they can become platforms of their own, as new services, like food delivery, are added in addition to transit]. This marketplace works because people are being augmented with new cognitive superpowers. For example, because of GPS and mapping apps, Uber and Lyft drivers don’t need a lot of training.

Agreed. But when the curtain rolls back we see that those superpowers have consequences: Those algorithms have bias built in.

That’s absolutely right. But I’m optimistic because we’re having a conversation about biased algorithms. We had plenty of bias before but we couldn’t see it. We can’t see, for example, that the algorithms that manage the workers at McDonald’s or The Gap are optimized toward not giving people full-time work so they don’t have to pay benefits. All that was invisible. It wasn’t until we really started seeing the tech-infused algorithms that people started being critical.

In WTF you talk about a specific out-of-control algorithm: the capitalist impulse to maximize profits regardless of societal consequences. The way you describe is reminds me of Nick Bostrom’s scenario of an AI machine devoted to making paper clips—because that’s its sole mission, it winds up eating up all the materials in the world and even killing those who would turn it off. Corporations whose sole justification is shareholder value seem to be working on a similarly destructive algorithm.

Yes, financial markets are the first rogue AI.

How do you roll back that particular AI?

I try to show [earlier cases of] how humans tried to manage their algorithms, by talking about [how Google improved] search quality. Google had some pretty bad patches where the spammers really had the upper hand, and they addressed it.

And that can be done to fix capitalism’s rogue AI?

Somebody planted the idea that shareholder value was the right algorithm, the right thing to be optimizing for. But this wasn’t the way companies acted before. We can plant a different idea. That’s what this political process is about.

Speaking of politics, it seems like another runaway algorithm has led us to a government controlled by people who don’t represent majority views.

I look at it through the long arc of history. You look at the long slow decline of the Roman Empire and see so many analogies—the turning away from literacy and science, the outsourcing of core functions of government to mercenaries effectively. We could go through some real shit before we turn around. We might not turn around at all. But I take hope from something that Tim Urban in Wait But Why calls “the human colossus.” He has this fabulous description of how Elon Musk moves this human colossus in a new direction—to show that it’s possible to go into space, to show that it’s possible to build a brain-machine interface—and then everybody else will follow along. The human colossus I’m most heartened by is the post-World War II period. We learned a lesson from the incredible convulsions after World War I where there was vast dislocation, as we punished the losers of the war. So after World War II they rebuilt Europe, and they invested in the returning veterans with the GI Bill.
As we learn from tech, though, algorithms need continual improvement. You don’t just set them in motion and leave them forever. The strategies put in place after World War II that worked for this period of 30 years have stopped working so well, so we came up with something else [which happened to create income inequality]. There’s a sense that things going wrong will lead to new strategies. And now that Trump has broken the Overton Window—

What’s that?

It’s this idea [named for the late think tank leader Joseph Overton] that there’s a certain set of things that are considered acceptable in public policy debate, and you just can’t go outside that window. And Trump has just done everything unthinkable. Because all bets are off, we are not necessarily going back to the old, tired solutions. I think it’s possible that we’ll shrug off this madness, and we will come back to saying we really have to invest in people, we really have to build a better economy for everyone. In China, they’re already doing that. China has recognized that its vast population is a possible powder keg and it has to take care of its people. That’s something we have not done. We’ve just been pushing down the ordinary people. China is also being more aggressive than any other country in rising to the challenge of climate change. So there’s two possibilities—we’re going to wake up and start acting the same way, or China will lead the world.
Reading your book I think I know who you’d like for our next president: Jeff Bezos. The book is full of Bezos love.

Well. Jeff and Elon [Musk] are probably the two entrepreneurs I admire most.

You can think of the book as an apology to Jeff. As a publisher, I originally bought the usual story, that Amazon would go the way of Wal-Mart—the more dominant it got, the more it would extract value for itself, squeezing down its suppliers. Jeff is a ruthless competitor, no question, but while Amazon has done a chunk of that, it has spent so much time trying to do more. I’m not sure that Jeff would make a great president, but he might.

You’d vote for him, wouldn’t you?

It would depend who he was running against but, yeah, I probably would.

You also praise Uber in your book. Do you think it’s possible to distinguish between the value of its service and the ethics of the company?

Uber is a good metaphor for what’s right and wrong in tech. Here we have this amazing new technology, which is transforming an industry and putting more people to work than worked in that industry before, creating great consumer surplus, and yet it has ridden roughshod over cities, and exploited drivers. It’s interesting that Lyft, which has been both more cooperative in general and better to drivers, is gaining share. That indicates there’s a competitive advantage in doing it right, and you can only go so far being an ass.
Let’s finish by talking about AI. You seem a firm believer that it will be a boon.

AI itself will certainly not take away jobs. I recently saw a wonderful slide from Joanna Bryson, a professor from the University of Bath. It referred to human prehistory and the text said, “12 thousand years of AI,” because everything in technology is artificial intelligence. What we now call AI is just the next stage of us weaving our intelligence together into a greater whole. If you think about the internet as weaving all of us together, transmitting ideas, in some sense an AI might be the equivalent of a multi-cellular being and we’re its microbiome, as opposed to the idea that an AI will be like the Gollum or the Frankenstein. If that’s the case, the systems we are building today, like Google and Facebook and financial markets, are really more important than the fake ethics of worrying about some far future AI. We tend to be afraid of new technology and we tend to demonize it, but to me, you have to use it as an opportunity for introspection. Our fears ultimately should be of ourselves and other people.

What’s the Matter With Applebee’s? – Eater

What’s the Matter With Applebee’s? – Eater

The experience of sliding into an overstuffed leather booth, hemmed in by walls decked in dubious Americana, the metal signs and pilfered taxidermy alluding to a time and place steeped in myth and wholly alien to the strip mall outside, while perusing a menu of oversauced fried hunks of protein and cheap carbs, all under the tawny haze of a poorly cloned Tiffany lamp, wasn’t quite universal. But it was common enough that the market, the great American arbiter of truth and beauty, blessed the suburbs from coast to coast — where so many of us were spawned and haltingly shepherded toward nominal adulthood — with thousands upon thousands of places in which to have that experience: The casual dining chain bloomed, almost like an onion you might say.

And now it’s dying, sort of. Because they’re terrible places, or because of millennials, or because of looming class warfare, or probably all of the above. Whatever the reasons, it should probably be less surprising that a monoculture as vast and mediocre as the suburban sitdown restaurant has contracted a terminal illness now slowly spreading from specimen to specimen, from Applebee’s to Ruby Tuesday’s to BW3 or whatever the fuck they’re calling Buffalo Wild Wings these days. More fascinating than the grinding demise of this corporate culinary hegemon, maybe, is the knowing, mournful soundtrack that we can’t help but provide with the collective gnashing of our teeth: Why do we still care so much about these places that we’ve since decided offer us such hollow fulfillment? —Matt Buchanan

[There are links to several articles about the decline of the suburban restaurant chains.  All the articles are depressing given the state of middle class America the chains represent.]

Today’s roads can’t handle today’s climate conditions | Smart Cities Dive

Today’s roads can’t handle today’s climate conditions | Smart Cities Dive

Dive Brief:

  • Engineers are using outdated temperature data — from 1964 to 1995 — to pick the right temperature-sensitive asphalt blends for use on roads today, Ars Technica reported, citing a new study in Nature Climate Change, which found the mismatch could raise road maintenance costs significantly.
  • In a study of nearly 800 asphalt roads built in the U.S. over the last 20 years, researchers found that 35% used an asphalt product ill-suited to current climate conditions. For one-quarter of those cases, the roads weren’t built to handle the high temperatures it experienced currently.
  • The researchers note that using an asphalt product even one grade short of what would be necessary could cut a few years off the road’s life, requiring repaving sooner than anticipated.

Dive Insight:

One in five miles of highway pavement was in poor condition in 2014 with urban roads (32%) worse off than rural ones (14%), according to the American Society of Civil Engineers. Poor roads don’t cost only those charged with their maintenance and repairs, however. That level of disrepair caused vehicle operators $112 billion in additional fixes and operating costs that year.

State and local governments paved roads in asphalt when prices for the material were low. Now, they are now rethinking that strategy as construction costs trend upward. To help manage those costs, at least 27 states have turned some of their asphalt roads to gravel, with most of that work occurring in the last five years, the ASCE reported.

Omaha, NE, is one municipality doing just that. The New York Times reported earlier this year that the city decided to convert some asphalt roads, including those in higher-end neighborhoods, to gravel after it struggled to fund repairs, such as filling potholes.

Montpelier, VT, has also made headlines for its de-paving activities. The city saved $120,000 by replacing asphalt on some run-down roads with dirt and gravel reinforced with a geotextile for stability rather than repaving, Wired reported.

A substantial number of de-paved roads are in rural areas that don’t see heavy traffic, according to a 2016 study from the National Cooperative Highway Research Program. Still, regardless of their location, drivers aren’t necessarily in favor of gravel roads, which generally lead to more wear on their vehicles as compared to asphalt or concrete surfaces.

However, it’s likely more state and local governments will look to de-paving as a way to manage road maintenance costs as questions continue around potential new long-term funding sources for infrastructure.

National Aquarium | Cephalopods: Arms or Tentacles?

National Aquarium | Cephalopods: Arms or Tentacles?

Cephalopods are a class of marine mollusks including the octopus, cuttlefish and squid. The name cephalopod means “head-foot” because they have limbs attached to their head, and these mollusks are well-known for their arms and tentacles. And while all cephalopods have arms, not all cephalopods have tentacles.

Tentacles are long, flexible organs found on invertebrate animals. They are important for feeding, sensing and grasping. Tentacles are longer than arms, are retractable and have a flattened tip that is covered in suckers.

Arms are similar to tentacles, but still distinctly different. Arms are covered with suckers that help with grasping food items. In addition, these arms are useful to attach to surfaces while resting.

The names may seem interchangeable, but when it comes to cephalopods, there’s a difference between arms and tentacles. An easy way to spot the difference is that arms have suckers along their entire length, while tentacles only have suckers at the tip.

This means that octopuses have eight arms and no tentacles, while other cephalopods—such as cuttlefish and squids—have eight arms and two tentacles.