11778 stories
·
90 followers

Why Renaissance Paintings Aren't as Green as They Used to Be

1 Comment
Comments

The Italian late-Renaissance painter Angelo Bronzino spent two years on his 1591 painting “Noli me tangere,” or “Christ the gardener.” The oil-on-wood painting, which was commissioned by a man who wanted to adorn his father’s funeral chapel, would make any dead father proud, depicting a beefy Christ and Mary Magdalene dressed in vivid blues and greens. But if Bronzino saw the painting now, he would probably be sorely disappointed. Over the past four centuries, the once-brilliant green paint has faded into a mucky, unrecognizable brown.

Noli me tangere,” which hangs today in a gallery of the Louvre, is one of many Renaissance paintings that features a copper-based pigment called verdigris. When fresh, its shade of bluish green is rare and luminous. But like many pigments popular in the 15th through 17th centuries, verdigris is toxic and unstable, Arthur DiFuria, an art historian at the Savannah College of Art and Design, explained in an interview with Copper.org, the website of a trade group that represents the copper industry. By the 19th century, verdigris had fallen out of fashion—mostly due to its poisonous nature—but no one ever figured out why the brilliant green pigment darkened so severely. Now, researchers in France have sleuthed the chemistry behind verdigris’s shadowy tendencies in a study in Inorganic Chemistry published in September, 2019.

In Jean Fouquet’s "<em>Pietà</em>," the dark coat of the man on the left was originally a bright green.
In Jean Fouquet’s “Pietà,” the dark coat of the man on the left was originally a bright green. Public Domain

To anyone living in the 21st century, it might not be obvious that Renaissance paintings were once much colorful than they look now. “If you look at the paintings of, say, Leonardo da Vinci, they are very, very dark,” says Didier Gourier, a chemist at the French National Centre for Scientific Research and an author of the study. “But they didn’t always look this way.”

Gourier knew that past researchers had speculated that light exposure and oxygen may have contributed to the darkening process, and he decided to analyze the chemical changes that took place in the verdigris. Though they had a plethora of paintings to choose from, the researchers selected two Renaissance paintings from the Louvre: Bronzino’s “Noli me tangere” and Jean Fouquet’s “Pietà.” Both works used plenty of verdigris that had sallowed over the years. Gourier took several samples, each smaller than a millimeter, and ran them through an electron microscope.

Presented with incredibly high-resolution images of the paint chips, they contrasted the color changes in verdigris sampled from the center of Bronzino’s painting against verdigris sampled right next to the frame, a shaded area that would have offered protection from light. Their suspicions were proven right when they found the frame-protected paint was far less deteriorated. When Gourier magnified a cracked paint sample from “Pietà,” he found that each crack had darkened, likely due to the diffusion of oxygen in the cracks. “The darkening is not systematic,” Gourier says. This inconsistency helps researchers pick out now-brown verdigris from originally brown paint, he says.

Microsamples taken from Bronzino's painting,<em>Noli me tangere</em> show vastly different levels of deterioration.
Microsamples taken from Bronzino’s painting,Noli me tangere show vastly different levels of deterioration. C2RMF, Inorganic Chemistry

To chemically confirm their theories, the researchers decided to recreate verdigris according to a medieval formula and see how they darkened over an accelerated time scale. “We had to speed up the darkening, because a painting in the Renaissance period would have taken several hundreds of years to darken,” Gourier says. “We calculated that 16 hours of LED illumination corresponds with several hundreds of years of illumination by museum light.”

Verdigris, technically known as copper acetate, has a simple recipe. Simply place metallic copper in vinegar and wait three or four weeks for the metal to react with the acid, producing blue-green copper acetate on its surface. (The Statue of Liberty appears blue-green for the same reason.) The researchers mixed the pigment with boiled linseed oil to make paint, as was the custom in the Renaissance. Gourier then placed the recreated verdigris on a thin sheet of glass to allow (simulated) centuries of light to pass through the sample. As if on cue, the gaudy verdigris darkened into muddled brown, just as the researchers expected.

Though Gourier intentionally selected two paintings with poorly-aged verdigris, the pigment can be detected a little more clearly on other famous works of Renaissance art. Chemical analysis is required to know for sure that a particular painting used verdigris (and it goes without saying that a permit to sample a 15th-century masterpiece is not easily acquired), but the charismatic green can be spotted in several works by Sandro Botticelli, such as the resplendently verdant “The Mystical Nativity,” which depicts—you guessed it—Christ and the Virgin Mary. DiFuria also suspects that Jan van Eyck’s green-sheened “Ghent Altarpiece” uses verdigris. The greens in these paintings are presumably less prismatic than they would have been several centuries ago, but they still, somehow, seem to glow.


Comments
Read the whole story
fxer
1 day ago
reply
Chemistry trigger
Bend, Oregon
Share this story
Delete

Restoring the Salvator Mundi

1 Share
Comments

I first saw the Salvator Mundi on April 27, 2005. A longtime friend, an art historian and dealer in Italian Old Master paintings, brought it to the apartment I shared with my husband, Mario Modestini. He had just received it from a New Orleans auction house and was hoping that I would agree to restore it. As is now well known, the Salvator Mundi was later recognized as a lost work by Leonardo da Vinci, and in 2017 it became the most expensive painting ever sold, bringing in $450.3 million at auction at Christie’s. But it was Mario, in the last months before his death, who first recognized the power of this disquieting portrait of divinity—a persona unseeing, all-knowing, eternal, pitiless, like the universe itself. After all that has happened since that day in April—the sale, the debates over authenticity, the accusations of manipulation—I often wonder what Mario would say about the doubters.

As my husband was, I am a restorer of Old Master paintings. I am good at it and treat my responsibility with the utmost gravity. The restoration of a badly damaged painting, especially when it is by a very great artist, always arouses criticism, not least from oneself. It is, unavoidably, an interpretation, like a musical performance except some of the notes are missing. All you have to go on are the well-preserved areas of original paint. These are sacrosanct.

While restoring the Salvator Mundi—work that I began in April 2005 and finished in September 2010—I made sure that my feeble retouches never masked these precious traces. To match the original paint, I built up the retouches in thin layers in the same sequence used by Leonardo. Fearful of covering any original paint, I used the tiny 000 sable watercolor brushes made by Winsor & Newton. Each brushstroke was carefully judged, based on knowledge of similar works and the formal structure of adjoining passages.

Even so, many of the gadflies who make their living on the fringes of the scholarly art world appear to believe that a restorer—in this instance, me—is capable of creating the Salvator Mundi. I suppose I should be flattered. One well-regarded art historian repeated this absurd notion in her soon-to-be published book on Leonardo. As Mario used to say, “She must be blind.”

The power of this disquieting portrait of divinity—a persona unseeing, all-knowing, eternal, pitiless.

It is now 500 years since the death of Leonardo and fewer than 20 paintings have been attributed to him. Many of them were not documented in his lifetime, and there are numerous copies by his followers. New attributions have always been contested and, understandably, must meet the highest standards. In 2008, after three years of examination, a majority of the National Gallery’s panel of experts accepted the Salvator Mundi, justifying its display in the museum’s 2011–12 exhibition.They responded to the magisterially painted blessing hand; the ringlet curls on the left, identical to those in Saint John the Baptist; the rendering of the hand refracted through the crystal sphere, with its accurately described inclusions; and the many-layered buildup of the flesh tones.

In 2017, the Salvator Mundi, Latin for “Savior of the World,” sold at auction for $450.3 million.

Since the sale, the painting has disappeared from public view and cannot speak for itself, leaving hacks and gossip-mongers to advance any tale or theory they choose. One of these is that the Louvre does not wish to include the painting in its exhibition on Leonardo—opening later this month—because its curators do not accept the attribution. I know with absolute certainty that this is pure fabrication. The reason for the Salvator Mundi’s absence from the Louvre show resides with the painting’s likely new owner, Saudi prince Bader bin Abdullah bin Mohammed bin Farhan al-Saud. (Most recently, the painting was scheduled to be unveiled at the Louvre’s Abu Dhabi outpost, before its display there, too, was postponed indefinitely.)

I cannot deny that my privileged relationship with the Salvator Mundi—in this post-truth era—has been difficult. But after the nonsense has run its course, that trance-like gaze, confined by the dimensions of the panel yet exploding from it, will remain: the most supreme representation of spirituality that has ever been painted. This is what my husband, Mario, saw; what transfixed me; and what I humbly sought to restore. —Dianne Modestini


Comments
Read the whole story
fxer
1 day ago
reply
Bend, Oregon
Share this story
Delete

WeWork pulls thousands of phone booths out of service over formaldehyde scare

1 Share

WeWork, the co-working empire once valued at $47 billion before reality struck, plunging the business and its investors into crisis, has another problem to add to its growing pile — one which doesn’t exactly reflect well on its core business of kitting out and maintaining modern working environments.

The problem is a safety concern affecting users of WeWork co-working spaces in the U.S. and Canada. Today the company emailed members in the regions to warn that around 1,600 phone booths installed at WeWork locations have been found to have elevated levels of formaldehyde — which it warns could cause health issues for people exposed to the gas.

WeWork blames the issue on a manufacturer of the booths.

The booths are provided in its co-working spaces for WeWork members to be able to take calls in private — given other common areas are shared by all users. 

“After a member informed us of odor and eye irritation, WeWork performed an analysis, including having an outside consultant conduct a series of tests on a sampling of phone booths. Upon receiving results late last week, we began to take all potentially impacted phone booths out of service,” it writes in an email to members.

Affected phone booths “are being taken out of service immediately, and will be removed from your location as soon as possible,” it adds. 

In addition to ~1,600 booths it has confirmed are affected, a further 700 booths are being taken out of service in what WeWork describes as “an abundance of caution” — i.e. while it carries out more checks — with the promise of a further update once it has concluded its tests. 

Members wanting to know which booths are safe to use in the meanwhile are told to contact the community team at their WeWork location.

WeWork also says alternative quiet spaces will be provided, such as in conference rooms and unused offices. 

Discussing the health risks of formaldehyde gas — a chemical which is used in various building materials –WeWork’s email warns: “Short-term exposure to formaldehyde at elevated levels may cause acute temporary irritation of the nose, throat, and respiratory system, including coughing or wheezing. These effects are typically transient and usually subside after removal of the formaldehyde source.

“Long-term exposure to formaldehyde, such as that experienced by workers in jobs who experience high concentrations over many years, has been associated with certain types of cancers. You can find additional information in this FAQ from the Occupational Safety and Health Administration.”

The email encourages any WeWork members with health concerns to contact a doctor.

A tipster who sent us the email reported experiencing a sensation of “burning eyes” after using the booths.

They also said several people in their team had experienced the same issue.

“Some complained that they felt nauseous after spending time inside the booths,” the tipster wrote. “I never felt that, but the burning eyes was 100% there for me several times. Scary stuff.”

Reached for comment, a WeWork spokesperson confirmed the formaldehyde issue, saying it’s taking “a number” of booths out of service at “some” locations in the U.S. and Canada — due to “potentially elevated levels of formaldehyde caused by the manufacturer.”

“The safety and well-being of our members is our top priority, and we are working to remedy this situation as quickly as possible,” it adds in a statement.

It is not clear exactly how many WeWork locations contain affected booths at this point.

Nor has WeWork provided more detailed information about how long members might have been exposed to elevated levels of formaldehyde — with its email merely suggesting some of the booths have been in place for “months.” 

“The potentially impacted phone booths have been installed over the past few months, exact timing varies based on location,” it writes.

Although clearly the level of exposure will vary from person to person depending on their use of the booths.

The company did not respond to a question asking whether any of its international WeWork locations are affected by the issue.

Read the whole story
fxer
1 day ago
reply
Bend, Oregon
Share this story
Delete

Researchers find just two plague strains wiped out 30%-60% of Europe

1 Share
Remains of human plague victims in a mass grave in Toulouse, France, dating back to the Black Death period.

Enlarge / Remains of human plague victims in a mass grave in Toulouse, France, dating back to the Black Death period. (credit: Archeodunum SAS, Gourvennec Michaël)

The Black Death ravaged medieval Western Europe, wiping out roughly one-third of the population. Now researchers have traced the genetic history of the bacterium believed to be behind the plague in a recent paper published in Nature Communications. They found that one strain seemed to be the ancestor of all the strains that came after it, indicating that the pandemic spread from a single entry point into Europe from the East—specifically, a Russian town called Laishevo.

Technically, we're talking about the second plague pandemic. The first, known as the Justinian Plague, broke out about 541 CE and quickly spread across Asia, North Africa, the Middle East, and Europe. (The Eastern Roman Emperor Justinian I, for whom the pandemic is named, actually survived the disease.) There continued to be outbreaks of the plague over the next 300 years, although the disease gradually became less virulent and died out. Or so it seemed.

In the Middle Ages, the Black Death burst onto the scene, with the first historically documented outbreak occurring in 1346 in the Lower Volga and Black Sea regions. That was just the beginning of the second pandemic. During the 1630s, fresh outbreaks of plague killed half the populations of affected cities. Another bout of the plague significantly culled the population of France during an outbreak between 1647 and 1649, followed by an epidemic in London in the summer of 1665. The latter was so virulent that, by October, one in 10 Londoners had succumbed to the disease—over 60,000 people. Similar numbers perished in an outbreak in Holland in the 1660s. The pandemic had run its course by the early 19th century, but a third plague pandemic hit China and India in the 1890s; there are still occasional outbreaks today.

Image of <em>Yersinia pestis</em> seen at 200× magnification with a fluorescent label.

Image of Yersinia pestis seen at 200× magnification with a fluorescent label. (credit: CDC/Larry Stauffer, Oregon State Public Health Laboratory)

"The second plague pandemic has arguably caused the highest levels of mortality of the three recorded plague pandemics," the authors wrote in Nature Communications. "It serves as a classic historical example of rapid infectious disease emergence, long-term local persistence, and eventual extinction for reasons that are currently not understood." And that makes studies of the genetic history of the bacterium behind the plague of great interest to epidemiologists, since genetic factors could influence the emergence and spread of such deadly pandemics.

Medieval doctors believed the disease spread via "bad air," or "miasmas." The real culprit is a bacterium called Yersinia pestis. While visiting Hong Kong in 1894 to study a plague outbreak there, a French scientist named Alexandre Yersin (who had studied under Louis Pasteur) extracted pus from a dead soldier’s swollen lymph node (bubo) and injected it into guinea pigs; all the guinea pigs died.

The bodies of a large number of dead rats around Hong Kong contained the same type of bacteria. Yersin concluded that Y. pestis was the culprit for the spread of plague. (It was discovered independently by a Japanese scientist named Shibasaburō Kitasato, but the microbe is named after Yersin.) Other experiments with plague-infected rats and fleas revealed that when an infected rat was introduced to a group of healthy rats, the healthy ones only became sick if fleas were present. Y. pestis proved to be so virulent that mice died after being infected with just three bacilli.

A variant strain of Y. pestis was likely also the culprit behind the Justinian plague, according to genetic analysis of the teeth of two German victims and the remains of Justinian plague victims in an ancient German burial site. It was largely the same strain as bacterial samples gleaned from London's plague pits. And recent studies of ancient Y. pestis DNA reconstructed from plague victims in southern France, Barcelona, London, and Oslo were identical.

For Maria Spyrou and her colleagues at the Max Planck Institute for the Science of Human History, this was clear evidence that a single strain spread throughout Europe during the Black Death. But where did it come from, and how specifically did it spread? Without genomic data from early outbreaks in Russia and only a limited number of published genomes for Y. pestis, making a definitive conclusion is difficult.

So Spyrou et al. set out to expand the selection of genomes from various time periods and locations, the better to study the early stages of the pandemic, as well as the genetic diversity that appeared in Europe after the Black Death. They were able to reconstruct plague genomes from the teeth of 34 victims culled from a collection of 180 teeth from nine sites, including two from Laishevo, which is part of Russia's Volga region. They also reanalyzed existing published data from the same time period.

A single strain did indeed prove to be the forerunner to all the strains from the second pandemic, although the team noted that earlier strains may yet be found in DNA samples from sites yet to be tested. Neither was there much genetic diversity between samples from victims during the Black Death, bolstering the view that Y. pestis found its way to Europe via a single entry point.

Once Y. pestis found a foothold in medieval Europe, it branched off into clades. Analysis of Y. pestis genomes from later in the second pandemic revealed two sister lineages. One appears to be responsible for the spread of Y. pestis eastward, since that lineage includes strains from 14th-century Bergen op Zoom, London, and the city of Bolgar, as well as some strains from Africa. The second, post-Black Death lineage showed an unusual degree of genetic diversity within local pockets in Germany, Switzerland, England, and France. Since there appear to be no modern descendants of this lineage, the authors surmise that it went extinct.

DOI: Nature Communications, 2019. 10.1038/s41467-019-12154-0  (About DOIs).

Read Comments

Read the whole story
fxer
1 day ago
reply
Bend, Oregon
Share this story
Delete

Judge grants mega-rich Sackler family reprieve from legal costs of opioid crisis

1 Share
Protestors hold up a banner while surrounded by empty prescription bottles.

Enlarge / PURDUE PHARMA, STAMFORD, Conn. - 2019/09/12: Members of P.A.I.N. (Prescription Addiction Intervention Now) and Truth Pharm staged a protest on September 12, 2019, outside Purdue Pharma headquarters in Stamford, over their recent controversial opioid settlement. (credit: Getty | Erik McGregor)

OxyContin-maker Purdue Pharma and its billionaire owners, the Sacklers, on Friday got a temporary reprieve from lingering court battles over their alleged role in fueling the opioid crisis. In exchange, they may have to be more forthcoming about what happened to all the OxyContin money.

US bankruptcy Judge Robert Drain temporarily halted state lawsuits against Purdue as well as the Sacklers—though only Purdue has filed for bankruptcy protections. In pausing the states' cases, Judge Drain cited Purdue's mounting legal expenses, which he noted is money that could otherwise go toward addressing the opioid crisis and its victims, according to The New York Times.

Purdue had sought a 180-day injunction on the state's cases, but Judge Drain's pause only lasts until November 6. In that shorter timeframe, he pushed the parties to try to talk out their differences. Those differences primarily hinge on whether the Sacklers are offering enough of their allegedly ill-gotten fortune to address the opioid crisis in thousands of lawsuits on the matter.

Purdue filed for bankruptcy in September as part of a tentative deal to settle more than 2,000 of them, which were mostly brought by state and local governments. According to Purdue's lawyers, the offered deal would be worth a total of $10 billion to $12 billion over time. Included in those figures is at least $3 billion from the Sacklers directly.

While hundreds of municipal governments and some states have agreed to the deal, about two dozen states and other plaintiffs are opposed. They argue that the Sacklers should pony up more of their fortune—and faster.

Billions

Purdue is estimated to have made more than $35 billion from OxyContin sales, and the Sacklers reportedly siphoned off as much as $13 billion of that into their own pockets.

Last month, New York's attorney general's office announced that it had tracked some of the Sacklers' wire transfers, which moved at least $1 billion of Purdue earnings into overseas accounts and obscure real estate entities.

To sweeten talks during the legal hiatus, Purdue and the Sacklers said it would share more financial information with the states. They also said they would not hide assets.

In a statement to The Wall Street Journal, the Mortimer D. Sackler and Raymond Sackler families issued a joint statement saying that they are "hopeful this process will enable all parties to reach a resolution."

Connecticut Attorney General William Tong, who has opposed the settlement deal, said in a statement that "We are disappointed" that the judge halted the cases, "but pleased that it is limited in time to less than 30 days. We will use this time to ensure that we get access to the Sacklers' financial information."

Read Comments

Read the whole story
fxer
1 day ago
reply
Bend, Oregon
Share this story
Delete

Lambda School is SV big bet on reinventing education but students say is a cult

1 Comment
Comments
  • Some of Silicon Valley's best and brightest believe that Lambda School is a new model for vocational education that could make student debt obsolete. 
  • The big innovation: Rather than paying up-front for tuition to its 9-month coding bootcamp program, Lambda School asks most students to sign income sharing agreements — where they pay a portion of their salary for the first two years after getting a job that pays $50,000 or more.
  • However, we spoke to former and current Lambda School student, who say that it falls short of that promise, with under-qualified instructors and an incomplete curriculum that requires them to rely on self-teaching and outside resources to complete the program. 
  • Those students say that the program is like a "cult," where they worry about criticizing Lambda School for fear of getting kicked out, and where they're encouraged by staff — including CEO Austen Allred — to respond to critical social media posts with positive testimonials. 
  • The Lambda School students say that when they voice concerns about the program or complain of harassment from fellow students, they get brushed off, ignored, or made to feel they were the problem.
  • Lambda School is a graduate of the famed Y Combinator startup accelerator program, and has attracted $48 million in venture capital from investors including GV (formerly Google Ventures), Stripe, and even Ashton Kutcher.
  • "While we can't respond to specific inquiries out of respect for privacy, we appreciate the concerns of any and all of our students. We are continuously aiming to be transparent on where we need improvement and the steps we are taking to address them," CEO Austen Allred said in a statement.
  • Click here for more BI Prime stories.

Erica Thompson has always had an interest in technology. Her father, a municipal transit operator, taught her the basics of programming, which she practiced while he built computers in his free time.  

She initially studied music education, but after her father had a heart attack, she decided that it might be time to pursue a career in programming. She had just wrapped up a software development course at a local college in Los Angeles when she saw a Facebook ad for Lambda School — an online coding bootcamp that requires no upfront tuition.

She decided to take a chance to hone her skills and make herself more competitive in the job market, without paying out of pocket. 

It's that very sales pitch that's driven Lambda School, based in San Francisco, to a position of prominence in Silicon Valley. A graduate of the famed Y Combinator startup incubator program (previous graduates include Airbnb and Dropbox), it's gone on to raise over $48 million from investors like GV (formerly Google Ventures), Stripe, and even Ashton Kutcher. The school boasts that there are nearly 3,000 students currently enrolled. 

What makes it unique from other coding schools is its income sharing agreement (ISA) model. Students sign a contract, agreeing to pay 17% of their income for two years when they get a job paying at least $50,000 a year, with a maximum payout of $30,000. It also offers a less-popular choice to pay a flat $20,000 in tuition, instead. 

To many, Lambda School represents a better way of thinking about higher education and vocational training, as Wired put it in an August headline: "Lambda School's For-Profit Plan to Solve Student Debt." And because students attend Lambda School remotely for eight hours a day, it's theoretically open to anybody, anywhere. The model has proven so appealing, other startups are following suit with their own ISA-based business models

Lambda School boasts of its successes, saying that graduates of the 9-month program go on to work for companies like Amazon, Google, or Microsoft. According to the Council on Integrity in Results Reporting, 60.9% of Lambda School students were employed 90 days after graduation, going up to 85.9% within 180 days. Graduates earn a median annual base salary of $60,000, according to that same study — although a Lambda School spokesperson puts that figure at $70,000, and notes that many graduates are in rural areas where average pay is lower.

What Thompson found, however, was that Lambda School was very different than what she hoped it would be. She says that she was brushed off by staff when she reported racist harassment from two of her classmates.

Not long after, Thompson says she was told that she was in danger of being removed from the program if she didn't hit certain goals. She says that she ultimately was kicked out of Lambda School, towards the end of the program, just days after raising her concerns directly with CEO and cofounder Austen Allred. 

"It seems that if anyone speaks up and is too critical of the program in any of the channels, they react as if the student is the problem, though they mandate feedback daily," Thompson told Business Insider. 

This is indicative of the general atmosphere at Lambda School, according to 5 former and current students, most of whom asked for anonymity for the sake of their careers. We also spoke to applicants and other people familiar with the Lambda School program.  

They say that while Lambda School pitches itself as a first step towards better job opportunities, the reality can be more underwhelming: The curriculum is lacking, the instructors are often under-qualified, and students are afraid to speak out in a culture described as being akin to a "cult," they say. 

"Lambda School is not worth the life it takes from you, and it's not worth the dollar amount you agree to pay them back," a former student said.

Lambda School did not make Allred available for an interview. In a statement, Allred said:

"While we can't respond to specific inquiries out of respect for privacy, we appreciate the concerns of any and all of our students. We are continuously aiming to be transparent on where we need improvement and the steps we are taking to address them."

"We adopted the ISA model to open up access to students from all walks of life and are constantly iterating on our curriculum and processes based on their unique experiences and feedback. At the same time, we're working to streamline those same feedback loops to make them as effective as possible. There's always room to improve, and we welcome any additional feedback on how we can continue to raise the bar."

'Trust the process'

Allred is known to many in Silicon Valley as a charismatic leader with a compelling personal background. He moved to San Francisco from Utah to break into the tech industry, and says he first lived in a car while his career got off the ground.

He's said that he decided to start Lambda School under the ISA model after being "taken aback" when speaking with someone who couldn't afford the $10,000 to attend a coding bootcamp. The goal, Allred has said, is to increase the economic opportunity for anybody, anywhere, who wants to build their own lucrative career in tech.

However, Lambda School students say that in reality, the program feels like a "cult" that attracts people down on their luck or otherwise in tough financial situations, and then locks them into an intensive program that ultimately leaves them on the hook to pay back thousands of dollars in their future wages.

"Lambda School is literally a cult," a former student said. "Cults are hard to leave. Cults play on your emotional vulnerability. Cults keep you mentally and physically exhausted so you can be more compliant…They're specifically targeting people who are vulnerable in hard-life situations."

To that point, 5 current and former Lambda School students tell Business Insider that they feel that they can't criticize or critique the program. 

A recent blog post from Allred highlights Lambda School's process for gathering feedback, where students are given constant opportunity to submit their thoughts, both directly to instructors and anonymously. 

However, the students say they don't feel comfortable airing any grievances: They're not only concerned about getting kicked out of the program, but also that they may end up blacklisted by companies like Nexient that are known to hire Lambda School graduates. 

Often, when students do bring up concerns about the school, they're just told to "trust the process," two sources say.

"It almost feels like gaslighting," a student said. "A lot of students have brought up similar concerns, and they're continually disregarded. There's a mantra they keep repeating: 'Trust the process of Lambda.'"

Troll defense

Allred, for his part, is known for personally responding to critics of Lambda School on social media. "People want to think that Lambda is a scam, because they want to believe the results we're producing are impossible," Allred recently told Wired.

It's common for students to spring to the school's defense, too. 

For example, when somebody posted to Twitter saying "Lambda School is trash," or when a Reddit user last year wrote a post saying "Is Lambda School really terrible?" users claiming to be Lambda School students chimed in with their thoughts. While those posts can sometimes contain critiques of the program, they're usually positive on Lambda School overall. 

"Your experience will vary based on who is your instructor and your PM, as with everything, some are better than others, but honestly, overall, I have had a really good experience, and I'm very happy with it," one commenter said.

However, these posts are at least sometimes made because Lambda School encourages students to defending the company's reputation in public, students say – with management, staff, and even classmates all known to encourage students to respond to any haters. 

A screenshot of Lambda School's main chatroom on Slack, shared with Business Insider, shows Allred himself thanking students for responding to a critical Twitter post. "Thanks guys for tweeting there, appreciate that," Allred wrote, as he shared a link to yet another Twitter post. 

'The curriculum is garbage'

The Lambda School students that Business Insider spoke with said that the actual curriculum has its problems, too. Expectations are unclear, they say, with constantly-shifting deadlines and classwork assignments that are themselves packed with software bugs. 

Some students say that to actually master the programming topics at hand, they had to use outside resources like Treehouse, Khan Academy or YouTube, because the Lambda School program itself wasn't sufficient. One former student goes so far as to say that it doesn't do enough to teach the fundamentals of computer science.

"Everyone knows the curriculum is garbage," another former student said. "They know it's not working. If you're keeping up, you either already had a foundation or you're self-teaching. The actual school is not effective at teaching. People are going outside to get what they need."

In general, many graduates who found jobs feel they would have been successful without Lambda School, though the school gets the credit for their success, two former students say.

Ultimately, some students say, they feel like they would have gotten the same or better education in coding with self-guided learning programs from places like Udemy or Khan Academy. 

"Lambda can do exactly the same as $10 course from Udemy," a former student said. "If you are able to get any resource on the Internet and spend the next few weeks actually reading through it and coding small projects, you're going to get a better experience than Lambda."

Students also give poor marks to the quality of teaching staff at Lambda School. Three current and former students complain that instructors are often themselves graduates of other coding bootcamps, with little real-world experience. One former student describes the instructors as "highly incompetent." 

Among the teaching staff is Ryan Allred, brother to CEO Austen Allred, who works at Lambda School as a data science instructor. However, his LinkedIn profile indicates that his experience in the tech industry consists of graduating from a web development bootcamp and working as an intern in the AI-related field of deep learning.

A Lambda School spokesperson defends his experience, and says he's "one of our higher rated instructors at the school." When Business Insider checked his LinkedIn profile after reaching out to Lambda School for comment, Ryan Allred's previous experience had changed from "Deep Learning Intern" to "Deep Learning Engineer."

Current and former students also say that Lambda School wouldn't be able to run without the labor of its team leads: students who are employed by the school to lead team meetings, fill out forms rating student performance, and conduct one-on-one meetings to review code. 

'Disorganized'

Former and current Lambda School students describe the program as "disorganized," with topics, projects, and even the length of the program itself seemingly changing at random. 

Starting in May, Lambda School extended the length of the program from 30 weeks to 9 months. A spokesperson says that existing students were given the option to either finish the program as scheduled, switch to the 9-month timeline, or else leave the program without triggering their ISA. No students stayed on the 30-week plan, the spokesperson said. 

However, three sources say that students weren't warned of the change at all and suddenly had their expected date of graduation extended by about a month with no warning. That presented a potential financial risk, the sources say, given that many simply don't have the time to work paying jobs while enrolled in the intensive program. 

This kind of disorganization appears to have led to at least one costly mistake for Lambda School: In August, Business Insider reported that Lambda School was facing a $75,000 fine for failing to obtain a key registration with educational authorities in the state of California, where it's headquartered.

At the time, Allred blamed Lambda School's former legal counsel for the decision not to apply for the registration in the first place. Failing to obtain that registration could endanger its ability to continue to operate, regulators told Business Insider. Currently, Lambda School's registration application is under review, authorities say. 

Read more: The hot Silicon Valley coding bootcamp Lambda School is paying a $75,000 fine for not registering properly with the state of California

Two people also say that the application process for Lambda School is confusing and inconsistent. Susan Money, from Michigan, said that she was unimpressed with the quality of a prerequisite screening class required to enter the program, and she never heard back from Lambda School after she completed it.

Another, Jacqueline Homan of Pennsylvania, who applied for the program hoping it would lift her out of poverty, said that she withdrew from that same class for medical reasons, and was told that she could re-enter at any time — but was bumped from Lambda School's Slack without notice in the interim period and couldn't raise Lambda School for help, even though she contacted the school to add her back.

Later, when Homan posted about her Lambda School experience on Quora, Allred responded, saying Homan's statements were false and that she was not accepted. Homan says she never received any rejection email.

"I basically got blown off," Homan told Business Insider. "If a program like that, that practically guaranteed job placement, if something like that isn't for someone like me, who the hell is it for?"

Diversity matters

Lambda School prides itself on a diverse student body. Because classes are held remotely, and because there's no upfront cost, it can accommodate students who might otherwise not have access to an education in programming.  However, current and former students say that they've been disappointed that Lambda School's teaching staff isn't diverse, in turn: most of the school's instructors are male or are not from underrepresented groups.

"Diversity is an important area of focus for the team," a Lambda School spokesperson said. "There's been little turnover on the instruction team, and the initial team was hired primarily on referrals from the founders' home state of Utah, not a very diverse state. As we've grown we've adopted a rigorous hiring process that has resulted in 5 of the last 7 instructor hires coming from underrepresented groups."

Still, students say, Lambda School can sometimes make for a learning environment that's uncomfortable for students from underrepresented groups, with staff doing little to intervene. Students recall instances of racist memes spreading through the Slack chatrooms, or when a white male student wore a Mexican sombrero to a presentation in front of the class.

On one occasion, a former student says, instructors started referring to each other as "Nazis." Lambda School says it was unaware of this incident and could not find a record of it on its internal Slack. 

"We take all forms of racism, sexism, and other discrimination very seriously," the spokesperson said. "Many students have been removed for violating our student code of conduct, which is primarily focused on ensuring a positive, safe learning environment for all students. We actively respond to inappropriate, unprofessional, and discriminatory content."

When students do report harassment, however, they're brushed off, ignored, or made to feel that they were the problem, students say. That was the case with Thompson, the student who says she was dismissed from the program after complaining of racist harassment. 

"They advertise the school as being for non-traditional students who may not be able to afford other routes into the industry, but those same students are also less likely to be able to get justice if something goes wrong," Thompson said. 

'You have to buy into it'

So, ultimately, is Lambda School worth it? That appears to be a matter of perspective. 

The $60,000 median base salary of Lambda School graduates, as reported by the Council on Integrity in Results Reporting, is slightly below the $65,000 median across all coding bootcamps according to Course Report, which studied programs such as Hackbright Academy, Hack Reactor, and Fullstack Academy. However, going with Lambda School's own figures of $70,000 makes for a more favorable comparison.

A Lambda School blog post says that it plans to share more data on graduate salaries and employment each quarter, starting early next year.

Furthermore, the income sharing agreement model doesn't necessarily mean that Lambda School is cheaper than its competitors in absolute terms. Both the $20,000 flat-rate tuition plan and the $30,000 cap on ISA repayments over the two-year period are well over the average bootcamp tuition of $13,584, also according to Course Report

To be sure, not all Lambda School students pay as much as $30,000. According to Lambda School's own math, if a student makes the minimum $50,000 annual salary that triggers the ISA, that student would be on the hook to pay $708.33 a month, totaling to nearly $17,000 at the end of the two-year repayment period.

A spokesperson points out that Lambda School's program is significantly longer than other programs of its like, and says that it includes more resources to help students get a job after graduation. The spokesperson said that the various coding education programs are very different, and it can "be like comparing apples and oranges." The spokesperson said that "we have always been open and upfront about the cost of our program but believe it is important to also be transparent about other school factors."

There still remains the question of cost, however, given that Lambda School's ISA means that students will be giving up a portion of their earnings after graduation for the two-year repayment period. 

"That is an affordability concern for students who are getting a job and will still have to make ends meet, pay for shelter, pay for food, and take care of medication and other life expenses," Joanna Darcus, staff attorney at the National Consumer Law Center, told Business Insider.

Still, despite these concerns, a former student says it's no surprise why so many of their fellow students and graduates jump to Lambda School's defense on social media and elsewhere. For them, Lambda School is a big, bold bet on their future success. That means that when things do go wrong, they may not want to admit it — even to themselves, the former student said. 

"If you're in Lambda School as a student, you have to buy into it," a former student said. "You told your friends and family and girlfriend and kids that you're going to become a software engineer. You don't want to look like a loser if you didn't make it. I think people put all their accountability on themselves for making it work when the school is failing them."

Got a tip? Contact this reporter via email at rmchan@businessinsider.com, Telegram at @rosaliechan, or Twitter DM at @rosaliechan17. (PR pitches by email only, please.) Other types of secure messaging available upon request. You can also contact Business Insider securely via SecureDrop.


Comments
Read the whole story
fxer
1 day ago
reply
Eh, kinds seems like a hit piece. There’s always gonna be dissatisfied students in any program at any school in any time period. The article took a very narrow view with “here are what some people who didn’t like it said”
Bend, Oregon
Share this story
Delete
Next Page of Stories