Author Archive: Martha Lee

Uber chief executive laments U.S. trade policy for lackluster IPO

Uber chief executive laments U.S. trade policy for lackluster IPO

WASHINGTON/NEW YORK (Reuters) – Uber Technologies Chief Executive Dara Khosrowshahi said on Tuesday the company’s initial public offering last month was hurt by U.S. trade tensions even as he downplayed recent volatility in the stock price.

Uber CEO Dara Khosrowshahi speaks at the The Economic Club of Washington in Washington D.C., U.S. June 11, 2019. REUTERS/Carlos Barria

The comments from Khosrowshahi continue the fallout from Uber’s IPO, which was the most anticipated U.S. listing since Facebook Inc seven years ago but has struggled since going public.

Uber priced shares in its IPO toward the bottom of its target range at $45 and the stock ended its first day of trading on May 10 down 7.6% at $41.57, even as the S&P 500 reversed losses to end in positive territory.

“The timing of our IPO was very much aligned with our president’s tariff wars, the same day,” Khosrowshahi said at an event at the Economic Club of Washington. “So I think we got caught up a bit in the market swirl. And there’s nothing you can do about that.”

Uber shares fell as low as $36.08 but have since rebounded and were trading at $42.21 on Tuesday, down 0.9% on the day.

Employees “have a six-month lockup so nobody in the company cares anyway what the stock prices is now – it’s a bunch of traders,” said Khosrowshahi. “It doesn’t really affect us… We work on building a great enterprise – the market will take care of itself,” he added.

Khosrowshahi addressed the announcement last week of the planned departure of its chief operating officer (COO) and chief marketing officer and he said he will not replace the COO as he organizes Uber “in a bit more of a business unit basis.”

He added they were “tough changes but I think are absolutely the right changes for the next three to five years.”

Khosrowshahi also has the option to buy potentially hundreds of thousands of shares in the ride-hailing firm at a price well below where the company went public, according to a regulatory filing for Uber’s IPO.

The options highlight how Khosrowshahi is at least partially protected from price swings in Uber’s stock.

Khosrowshahi has options on 750,000 Uber shares that vest gradually and can be exercised at $33.65 per share.

Uber first granted the options on Sept. 5, 2017 with an exercise price of $41.65 per share but this was changed on May 8, 2018, according to the filing. The details were reported earlier on Tuesday by The Information.

An Uber spokesman said the repricing of Khosrowshahi’s option grants was made to shorten the timeline over which they were granted to seven years from 10 years in conjunction with a bonus plan dependent on the company being acquired for or trading at $120 billion. Uber went public at a valuation of $82.4 billion..

There was no material change in the grant’s total value, according to the company’s filing.

“I believe in performance-based pay,” Khosrowshahi said Tuesday when asked about his compensation. “If I do very well for shareholders … that’s a circumstance where I want to be rewarded.”

Khosrowshahi joined Uber in 2017 to replace company co-founder Travis Kalanick who was ousted as CEO.

He was tasked with getting the company ready for its IPO, and convincing investors that he had successfully changed Uber’s culture and business practices after a series of embarrassing scandals over the last two years.

Reporting by David Shepardson in Washington and Joshua Franklin in New York; Editing by Nick Zieminski

Read More

Michelberger Hotel / Jonathan Tuckey Design

Michelberger Hotel / Jonathan Tuckey Design

Michelberger Hotel / Jonathan Tuckey Design

Michelberger Hotel / Jonathan Tuckey Design, © Philip Obkircher

© Philip Obkircher


© Philip Obkircher


© Philip Obkircher


© Philip Obkircher


© Philip Obkircher






+ 16



  • Architects

  • Location

    Warschauer Str. 39-40, 10243 Berlin, Germany

  • Category

  • Lead Architect

    Rob Leechmere

  • Project Year

    2018

  • Photographs

  • Manufacturers

    Loading…


© Philip Obkircher

© Philip Obkircher

Text description provided by the architects. The hotel occupies a former light industrial building that dates from 1903. It is located within the Berlin district of Friedrichshain, an area formed around manufacturing and built when the railway and waterworks were opening in the late 19th Century.


© Philip Obkircher

© Philip Obkircher


Hotel Drawing

Hotel Drawing


© Philip Obkircher

© Philip Obkircher

The building interior is made up of a single open-plan floor plate built around a glazed brick courtyard and traversed by a grid of curved concrete beams. The floor is then surrounded by 3-meter-high windows that look out onto the rapidly changing industrial surroundings.


© Philip Obkircher

© Philip Obkircher

There was an intention to layer domestic-scale rooms and materials over the Michelberger’s original features, without concealing the vast open floor-plate with its enormous windows and sculptural beams. This approach creates a visible tension between the building’s past and what it will become.


© Philip Obkircher

© Philip Obkircher

The differing spaces within each room are crowned by wooden palisaded frames that nod to the articulation of the S-Bahn signal box outside. They define a sense of enclosure and create a dialogue in scales between the divisions in the room and the city body. In contrast to the arborescent upper portions of the rooms, a series of datums have been introduced into the lower level.


© Philip Obkircher

© Philip Obkircher

The material, textural and chromatic datums attempt to describe the rooms in terms of; ground, plinth, firmament, and canopy. A range of fine furniture; desks, tables, chairs, beds, and benches have been designed by the practice. A proportion of these has been made from Panzerholz, a sheet timber material formed under high compression.


© Philip Obkircher

© Philip Obkircher

The natural resins secreted by the wood during manufacture, give the material an aged and lasting appearance. The fine profile of this material contrasts with its immense weight and strength, producing furniture which is both very delicate and sturdy.


© Philip Obkircher

© Philip Obkircher

View the complete gallery

Project location

Location to be used only as a reference. It could indicate city/country but not exact address.

Cite: “Michelberger Hotel / Jonathan Tuckey Design” 21 May 2019. ArchDaily. Accessed .

想阅读文章的中文版本吗?

© Philip Obkircher

米赫尔伯格酒店 / Jonathan Tuckey Design


Did you know?

You’ll now receive updates based on what you follow! Personalize your stream and start following your favorite authors, offices and users.

Read More

World’s First 360-Degree Infinity Pool Is Set to Open in London

World’s First 360-Degree Infinity Pool Is Set to Open in London

UK-based swimming pool manufacturer Compass Pools has announced plans for the world’s first 360-degree infinity pool. Titled Infinity London, the pool would contain 600,000 liters of water, and would sit atop a 55-story hotel.

According to Compass Pools’ plan, the pool would feature transparent walls and floor, offering views across London’s skyline and down into the hotel below. These walls would be constructed from cast acrylic, a material chosen to make sure that the material is completely clear.

The main issue with designing the pool was how to guarantee entry or exit from the pool without compromising the view. Speaking about this element of the design, Compass Pools’ technical director and designer Alex Kemsley explained that “normally a simple ladder would suffice, but we didn’t want stairs on the outside of the building or in the pool as it would spoil the view – and obviously you don’t want 600,000 litres of water draining through the building either. The solution is based on the door of a submarine, coupled with a rotating spiral staircase which rises from the pool floor when someone wants to get in or out.”

Take a look at early mock ups of how the pool might look in the gallery above. Currently, no contractors have been confirmed but building work could begin as early as 2020.

In other design news, IKEA and Olafur Eliasson have a new project.

What to Read Next

NikeTown London Adds Plus-Size & Para-Sport Mannequins to Encourage Inclusivity in Sportswear

More Swoosh stores to follow.


A-COLD-WALL* Teases Diesel Red Tag Collaboration

Set to arrive in late 2019.


'Fantastic Four' MCU Reboot Rumored for 2022

Looking forward to the revamped Invisible Woman, Mister Fantastic, the Thing, and the Human Torch?



Liverpool Reveals Blue & White 2019/20 Away Jersey

Referencing the club’s first European Cup win.


Netflix Confirms Fifth & Final 'Lucifer' Season

The streaming service promises a “big finish.”


Ed Sheeran Released His Own Heinz Ketchup for National Ketchup Day



More ▾

Read More

This AI tool is translating 2,000 African languages in a bid to boost local economies

This AI tool is translating 2,000 African languages in a bid to boost local economies

A digital platform called OBTranslate that aims to translate more than 2,000 African languages to enable rural dwellers to gain easy access to global markets has been launched.

According to its creator, 63 per cent of the population in Sub-Saharan Africa do not have access to global markets because of language barriers.

“Over 52 native languages in Africa have undergone language death and have no native speakers,” said Emmanuel Gabriel, founder of Germany-based OpenBinacle, the creator of OBTranslate, which was launched this month. “OBTranslate can close communication gaps on the continent.”

“In the next five years, we hope to acquire thousands or millions of users to take up translation tasks on OBTranslate.”

Emmanuel Gabriel, OpenBinacle

The innovation resulted from an earlier messaging app that was built in 2017 to allow interaction in real-time translation of 26 African languages, but led to inaccurate outputs, Gabriel admitted.

“We were very frustrated about the messaging app, and as a result we didn’t want to come into the market with a bad product,” added Gabriel. “We decided to embark on building our own computer-assisted translation and machine learning platform and this gave birth to OBTranslate.”

According to Pangeanic, a global translation company based in Spain, a computer-assisted translation tool converts texts into smaller and translatable segments to facilitate quick and accurate translation.

“The segments can be recalled later on and thus the translator ensures that the terminology and writing style of the original is followed,” Pangeanic explains on its website. “It also provides savings when the material that needs translation is similar to previously translated material – you only pay for part of the sentence that has changed.”

Gabriel added: “We created OBTranslate with an innovative business model to guarantee that the tool is sustainable to pay everyone whose translations are very accurate. In the next five years, we hope to acquire thousands or millions of users to take up translation tasks on OBTranslate.”

Gabriel believes the platform could make a positive contribution to Africa’s economy and job creation efforts. However, because some technical issues still have to be resolved, the team has not yet enabled real-time translation by public users.

“We are working hard to ensure that when people present training materials in African languages … they don’t necessarily tell the machines what to look for. The system [should] find patterns themselves such as contextual clues around the source sentence,” he explained.

OpenBinacle has received infrastructure worth US$100,000 from French cloud computing company OVH, as well as Amazon and Google, and are exploring funding opportunities to refine OBTranslate, Gabriel added.

Bukunmi Seweje, director of operations at Compucode Limited, Nigeria, believes that  OBTranslate could enable seamless business relationships in Africa.

But to make the innovation more user-friendly and accessible, Seweje said it should be integrated into existing communication platforms such as WhatsApp.

Read More

This magical metamaterial could revolutionize car safety and save lives

This magical metamaterial could revolutionize car safety and save lives

By Jesus Diaz3 minute Read

Scientists at the University of Washington have created a new shock absorption metamaterial that uses origami to completely absorb a hard impact and transform that crushing force into a gentle pull. The findings have potential implications in everything, from delivering packages via drones to landing spaceships.

According to one of the research authors—UW associate professor of aeronautics and astronautics Jinkyu Yang—”if you were wearing a football helmet made of this material and something hit the helmet, you’d never feel that hit on your head.” In fact, by the time the impact energy reaches you, he says, it has been transformed from a crushing push to a light pull.

[Image: courtesy University of Washington]

When you see it in action, it looks like magic. The idea of using a shape that can transform pushing forces into pulling forces is impossibly counterintuitive—something the researchers admit themselves.

The secret is the metamaterial’s geometry. “Metamaterials are like Lego,” Yang said in a press release. “You can make all types of structures by repeating a single type of building block, or unit cell as we call it. Depending on how you design your unit cell, you can create a material with unique mechanical properties that are unprecedented in nature.”

The UW research—published today in the journal Science Advances—got inspiration from origami to create 20 of these flexible unit cells, using a laser-cutting plotter to create physical models of a geometric shape they developed using computer simulations.

Then they put together the segments in a long truss. Each of the segments in the metamaterial then acts as a “folding crease,” which has the capability of softening an impact as it travels through the truss. In fact, as the shock advances through each segment, the segment before it bounces part of the energy back, pulling the next segment until the chain eventually dissipates the push force generating a soft draw instead.

Scientists then tested their initial computer models with the physical model by applying a compression force, recording the behavior using six GoPro cameras filming in slow-motion. And indeed, it worked exactly as the simulation predicted, turning compression forces into pull forces.

In this computer simulation, red represents the pushing force shock wave. The blue shows the pull force. You can see how the red transforms into blue as it travels through the metamaterial. [Image: courtesy University of Washington]

The way the unit cells fold is crucial, according to research coauthor Yasuhiro Miyazawa, who is completing an aeronautics and astronautics doctorate at the University of Washington: “[The origami] unit cell softens the force it feels when someone pushes on it, and it accentuates the tension that follows as the cell returns to its normal shape.”

The applications are countless: “Impact is a problem we encounter on a daily basis, and our system provides a completely new approach to reducing its effects,” Yang said.

In the future, the system could be used to soften the landing of packages delivered by drones, if you built the origami segments right on the packaging itself. The system could also be used in car bumpers to soften the impact of car accidents and, according to Yang, save lives.

Yang also points out that you could also make the same structures out of composite materials for greater strength and larger objects. You could actually apply this technology to landing systems, the team says, like the ones needed to send rovers to Mars or providing soft landing for SpaceX spaceships without having to use so much energy in the return.

Read More

When Will Quantum Computers Outperform Regular Computers?

When Will Quantum Computers Outperform Regular Computers?

Any day now, quantum computers will solve a problem too hard for a classical computer to take on. Or at least, that’s what we’ve been hoping. Scientists and companies are racing toward this computing milestone, dubbed quantum supremacy and seemingly just beyond our reach, and if you’ve been following the quantum computing story, you might wonder why we’re not there yet, given all the hype.

The short answer is that controlling the quantum properties of particles is hard. And even if we could use them to compute, “quantum supremacy” is a misleading term. The first quantum supremacy demonstration will almost certainly be a contrived problem that won’t have a practical or consumer use. Nonetheless, it’s a crucial milestone when it comes to benchmarking these devices and establishing what they can actually do. So what’s holding us back from the future?

“We’re about to cross over into a world where we’re doing something with quantum devices that we couldn’t do classically,” John Preskill, professor of theoretical physics at the California Institute of Technology who devised the term “quantum supremacy,” told Gizmodo. “We’re at a pivotal stage.”

You might first wonder what a quantum computer is—or, fundamentally, what a computer is. Computers are devices that abstract data and store it as inputs, which they manipulate via a system of instructions and mathematical algorithms. Typically, the data is stored as manipulatable bits, two-choice physical devices, allowing systems of these bits to produce some desired output. On quantum computers, algorithms are mapped onto a different kind of architecture; instead of bits, there are two-choice devices called qubits that obey the weird rules of quantum mechanics.

Each qubit is sort of like two-sided dice that you can load with an iron piece to adjust the probability you’ll roll either side. Performing a quantum calculation is like rolling the dice. But you can entangle the qubits, which is like magnetizing the iron pieces so that the dice must be treated as a single multi-sided dice with its own set of probabilities. This might lead to interference—making certain sides of the dice more likely to roll, and other sides less likely. Quantum calculations apply gates to these dice—pulses of energy that further influence the position of the weight inside the dice and change what you’ll roll. Applications involve mapping some piece of information to each side, and often require many dice rolls in order to achieve interesting results.

Scientists and technology companies are pursuing quantum computers both for their innate scientific interest as envelope-pushing experiments and for the way they might impact artificial intelligence, cybersecurity, and healthcare. Presumably, there are some algorithms that would work faster running on this quantum architecture than they would on a classical computer, most notably Shor’s algorithm, which could factor extremely large numbers faster than classical computers can. An algorithm that can quickly factor numbers is important because most of our modern-day encryption is based on the premise that classical computers can easily multiply two numbers together but take an unreasonably long time to factor them back apart. A computer that could run Shor’s algorithm would render this encryption strategy unsecure. Others hope that quantum computers will find applications in artificial intelligence, such as in quantum neural networks, or help solve chemistry problems, like finding new drugs to cure diseases, much more efficiently than classical computers.

But a quantum computer worth getting excited about—or worrying about—should be better than regular computers doing the same task. That’s why scientists and companies, most notably Google, list quantum supremacy as a key milestone for their devices.

Proposals for quantum supremacy generally follow the same premise. Set up complicated, random quantum circuits and measure the values. You get a lot of answers. Now, use a statistical test to ensure that the experiment was done properly using these outputted answers. Theoretical physicists think that a quantum computer would demonstrate supremacy on such a task—basically, as the problem increases in complexity, the amount of time it takes to calculate the problem would increase at a higher rate for a classical computer than a quantum computer.

From a business standpoint, the task seems contrived. Quantum supremacy demonstrates that a quantum computer is better at being a quantum computer than a classical computer is. That’s not something you can cure diseases with.

But from a theoretical perspective, it’s profound, Bill Fefferman, assistant research professor at the University of Maryland and research scientist at the National Institute of Standards and Technology, told Gizmodo. There’s a hypothesis called the Church-Turing thesis stating that any computer problem can be solved with an abstract kind of computer devised by mathematician Alan Turing in 1936. This theoretical computer simplifies all computing problems to symbols on tape. Then, there’s the extended Church-Turing thesis—that no practical model of computing could solve tasks significantly faster than one of these Turing machines. Theoretical computer scientists gave strong evidence that the extended Church-Turing thesis was wrong in the early 1990s. And a machine that achieves quantum supremacy would provide experimental evidence against the thesis. It would demonstrate that there really are computer problems for which a supercomputer wouldn’t be the most efficient way to calculate them, and for which a computer based on a different architecture, a quantum computer, would be.

Quantum supremacy, from a scientific perspective, is meant to give scientists a concrete way to determine what quantum computers will and will not be useful for, and to compare them against classical computers. Up until the early 1990s, theoretical computer scientists devised contrived problems for quantum computers, and useful tasks, like Shor’s algorithm, came later. “Instead of saying ‘you guys are spending billions of dollars to implement this contrived problem,’ the answer is that we have to first build the foundations,” Fefferman said.

It’s also not inherently useless, as the experiment could make quantum computers useful random number generators, which find applications in cryptography, gambling, simulations, and elsewhere.

But how do you actually hit that milestone? Google has enlisted the help of NASA in its quest to be first, MIT Technology Review reported last year. The team is building and testing a chip with what they hope will be enough good qubits to demonstrate quantum supremacy. There are researchers studying these same quantum supremacy problems on classical computers to compare, as well as working on computational theory to ensure rigorous proof that supremacy was achieved.

With the considerable resources and minds of NASA, Google, IBM, and other organizations on the case, you might wonder what’s taking so long. For now, the largest commercial quantum devices have around 20 qubits, though IBM, Google, and IonQ are testing 50-, 72-, and even 160-qubit devices, respectively. But every step of building and operating a quantum computer is hard. Instead of silicon transistors on microchips, scientists must create their devices either out of lasers that trap individual atoms, superconducting material that conducts current without resistance that demonstrates manipulatable quantum properties, or other potential architectures. This often requires holding the processor at nearly absolute zero—the temperature at which particles have the minimum possible heat. Controlling this system proves incredibly difficult, since a modicum of energy from the outside environment could cause the qubits to collapse into very expensive, very regular bits.

On these limited systems, researchers can only perform a handful of quantum operations, or “gates,” before the quantum state collapses. Entangling too many qubits will make the system collapse. Each additional qubit makes the machine twice as complex. Electromagnetic pulses that control the system must be perfectly engineered.

At the same time, quantum computer scientists aren’t just trying to beat classical computers simulating quantum computers. They’re trying to beat any possible workaround that someone programming a classical computer can come up with, which is harder to prove. And researchers will need to somehow verify the result that the quantum computer calculated, even though they would have just performed a calculation that a non-quantum computer can’t do.

“I bet someone is going to declare [quantum supremacy] soon,” Graeme Smith, assistant professor at the University of Colorado, Boulder, told Gizmodo, “But there will be questions about whether they achieved it because of the fact that it’s hard to verify.”

Perhaps you’re noticing a pattern here. No one has achieved quantum supremacy yet because it’s hard.

IBM’s scientists are working on a perhaps easier-to-achieve task. They’re trying to demonstrate “quantum advantage.” The difference is subtle. Quantum supremacy means that a quantum computer can perform calculations that a classical computer cannot in a reasonable amount of time. Quantum advantage just means that the quantum computer can beat the classical computer at some calculation, even if it’s just a little better. Some researchers have devised mathematical proofs of cases where quantum computers are always better than classical computers running the same algorithm. But in this case, the classical computer was given a similar limitation to one of the core present-day quantum computer limitations: it could only perform a few operations at a time, like a qubit, which can only perform a few operations until it collapses.

Quantum advantage has one leg over quantum supremacy, though. Quantum supremacy is a high bar to achieve, but if the industry is solely looking for a faster algorithm, then these quantum advantages might find quantum computers a more general use in industry sooner.

For companies like Google and IBM, these terms are tossed around with a heavy dose of PR. “They’re both trying to build programmable quantum computers,” Aram Harrow, associate professor of physics at MIT, told Gizmodo. “Google will say that the goal is supremacy, IBM will say that the goal is advantage. That’s not going to lead to a big difference in the hardware that gets built.”

Ultimately, when a company building a quantum computer inevitably announces that it has achieved “quantum supremacy” or “quantum advantage,” it won’t be a sea change for the industry. They’ll still be referring to relatively small, error-prone devices—what researchers call NISQ, or “noisy intermediate-scale quantum” machines. These machines will still face the same limitations they did before a supremacy or advantage device existed, such as the short amount of time qubits can stay quantum or lower number of calculations a qubit can do before losing its quantum nature.

“Quantum supremacy is a stepping stone for us to move forward in order to solve more interesting problems,” Mekena Metcalf, postdoctoral researcher at Lawrence Berkeley National Laboratory, told Gizmodo. But for quantum computers to become the code-cracking, molecule-simulating devices of the future, there are specific goals on the horizon. “It requires orders of magnitude more qubits and long gate depth,” she explained—qubits that can do more calculations before losing their quantum behavior.

Reaching that state will require better hardware, including more precise optics for quantum computers based on laser-trapped atoms, Sara Mouradian, postdoctoral researcher at UC Berkeley, told Gizmodo. Those working on superconducting quantum computers are hoping to see improvements to the system’s wires and better control overall. Both systems will need to find ways to vastly increase in scale and size, which is not as easy as adding more bricks to a Lego tower. Quantum computers will also require error correction, or storing the information from a single qubit across multiple entangled physical qubits in order to correct for possible errors.

The NISQ-era devices are still boundary-pushing tools for studying quantum physics, and perhaps they’ll find useful applications in the near term, whether they demonstrate “quantum supremacy,” “quantum advantage,” or even just “quantum usefulness.” There are plenty of other quantum devices in the works as well, like sensors and cryptographic tools, that might find applications sooner.

But, for startup funding’s sake, hopefully scientists and technologist demonstrate quantum supremacy (or advantage) and find useful quantum computer applications soon.

“What looms over the field, particularly on the commercial side, is all of these companies investing and building systems—but if they can’t come up with a useful application in 10 years, what happens then?” Preskill said. “Is there going to be a quantum crash because people felt that the potential wasn’t realized?” At least for now, the U.S. government has passed a bill injecting money into this sector in order to train more scientists and transfer knowledge to the industry.

Quantum supremacy is on the horizon, and the pursuit of it is continuing to push scientific progress in fundamental and profound ways. But proving quantum supremacy for one problem won’t bring quantum computers much closer to your own desk. It’s crucial that we continue to emphasize the uncertainties in the field, especially when it comes to near-term potential.

Quantum computers highlight that science and technology have different goals, and provide two very different lenses through which to understand quantum supremacy. Technology can feel like an endless march toward some better product. But science is slower, unpredictable, and often more rigorous—it requires that people cover all the bases in order to understand how these groundbreaking new devices actually work before we can claim that quantum computers are actually superior.

This story has been updated to clarify the relationship between quantum supremacy and the extended Church-Turing Thesis.

Read More

The Faults Beyond Our Algorithms

The Faults Beyond Our Algorithms

The offline roots of online radicalization.

Ross Douthat

Image

CreditCreditGetty Images

In the last month there have been two big profiles, one in this newspaper and one in Washingtonian, of young men who were radicalized into some version of the alt-right and then later abjured far-right politics.

The Washingtonian profile is a mother’s first-person account of watching her 13-year-old son join an extremist internet community and then find his way back out. The Times profile, by my colleague Kevin Roose, follows a 20-something college dropout who has turned to progressive politics after a sojourn as a “tradcon.” Both pieces locate their subjects in a larger sociology, but both are especially concerned with the role of YouTube in accelerating radicalization, and in particular with the way its algorithm, intended to reward videos that prompt intense engagement, is creating a pathway into extremist content by rewarding provocation and letting content purveyors essentially monetize the rabbit hole.

Since I might be described (even in the comment thread below this column, who knows?) as some sort of “tradcon” myself, I have some doubts about categorizations that purport to define the moment when respectable conservatism becomes dangerous extremism. Nonetheless I agree with several of the profiles’ anti-YouTube premises: Education-by-algorithm is a bad idea; Silicon Valley has monetized amoral slippery slopes in all kinds of arenas (with pedophilia being the latest grim example); and the stew of racism and conspiracy theory at the far-right edges of social media is an example of where one such slope can lead.

That said, the profiles are rich enough to illustrate a complexity that can’t just be reduced to algorithms run amok. When it comes to explaining the phenomenon of right-wing populism, liberals are likely to argue both that the populist era has exposed a darkness always present at the heart of conservative politics and that a toxic, post-truth new-media ecosystem has greased the skids for President Trump, Brexit and the rest. In the case of white-identitarian media personalities popular on YouTube, the two arguments converge.

But in the profiles you can also see clear illustrations of two arguments that a certain kind of conservative — the kind who locate their politics somewhere between the Trumpians and the Resistance — is likely to offer in response to liberalism’s explanations. First, that populism is a reaction to the breakdown of community outside the liberal metropole, a breakdown that the fiscally conservative-socially liberal “centrism” of our leaders has worsened or left unaddressed. And second, that far-right personalities attract people by offering an escape from the airlessness of liberalism, a chance to rebel against its cultural hegemony and increasing ideological conformism.

Caleb Cain, Roose’s subject, illustrates the first argument, to the extent that he seems like a character from the pages of Charles Murray’s “Coming Apart” or Tim Carney’s “Alienated America” or J.D. Vance’s “Hillbilly Elegy,” to cite just the most prominent right-leaning authors who have written about lower-middle-class decay:

The internet was an escape. Mr. Cain grew up in postindustrial Appalachia and was raised by his conservative Christian grandparents. He was smart, but shy and socially awkward, and he carved out an identity during high school as a countercultural punk. He went to community college, but dropped out after three semesters.

Broke and depressed, he resolved to get his act together. He began looking for help in the same place he looked for everything: YouTube.

In other words, a story about the internet is also a story about the attenuation of non-online forms of community, about fatherlessness and family breakdown, about the travails of the “some college” demographic, about the social crisis of coal country and the Rust Belt. And it’s therefore an example of why attempts to wall off explanations for populism into mutually exclusive “economic anxiety” and “racism” categories are so futile: There is an integrated relationship between socioeconomic conditions in downscale America and the appeal of redpilled YouTube, between changing personal and material conditions and the appeal of virtual white-identitarian “community.”

There is also an integrated relationship between how elite liberalism has evolved and the kind of reactions it provokes. Here is the opening act, from the Washingtonian profile, of the suburban-D.C. episode that sent its 13-year-old subject into the online wilds in search of support and solidarity:

One morning during first period, a male friend of Sam’s mentioned a meme whose suggestive name was an inside joke between the two of them. Sam laughed. A girl at the table overheard their private conversation, misconstrued it as a sexual reference, and reported it as sexual harassment. Sam’s guidance counselor pulled him out of his next class and accused him of “breaking the law.” Before long, he was in the office of a male administrator who informed him that the exchange was “illegal,” hinted that the police were coming, and delivered him into the custody of the school’s resource officer. At the administrator’s instruction, that man ushered Sam into an empty room, handed him a blank sheet of paper, and instructed him to write a “statement of guilt.”

No one called me as this unfolded, even though Sam cried for about six hours straight as staff members parked him in vacant offices to keep him away from other students. When he stepped off the bus that afternoon and I asked why his eyes were so swollen, he informed me that he would probably be suspended, but possibly also expelled and arrested.

… In an out-of-body moment, I imagined that this very episode would be cited by some future cultural critic on the limits of liberalism …

And yes, perhaps too predictably, here I am citing it — as a particularly vivid example, from inside the D.C. bubble, of how progressivism fosters alienation and potential rebellion long before any sinister Silicon Valley algorithms get involved. And, going further, as an example of how the alt-right’s promise of “secret knowledge” about gender, race and sexuality finds a ready audience among young people primed by aspects of their own progressive education to suspect that the system is enforcing ideology rather than imparting truth.

Which, let me stress again, doesn’t mean that the far-right YouTubers are actually good or the algorithms actually benign. Nor does it mean that conservatives reacting to liberal critiques of YouTube should automatically leap to the defense of either — on the assumption that as goes the YouTube success of Stefan Molyneux, so goes the non-racist right.

Yes, it’s understandable for conservatives to worry that if Silicon Valley censors the likes of Molyneux, it will end up censoring them. It’s sensible for them to join parts in the left in worrying about the concentrated power over information that the stewards of social-media platforms enjoy. And it’s necessary for them to recognize that the influence of redpillers and white-identitarians reflects their own failure, across the decades of movement-conservative institution building, to create something that seems more compelling to fugitives from liberalism than the Spirit of the Reddit Thread.

With all that said, though, a humane conservatism should still be able to thrive in a world where white nationalists have trouble monetizing their extremism, in which YouTube algorithms are built to maximize something other than addiction. And the endpoint of both the Times and Washingtonian profiles, in which personal contact with real-life alt-righters and counterarguments from online left-wingers suffice to pull the main characters away from the far right, suggests that the current Very Online alternative to liberalism is not necessarily built to last.

But if conservatives would be unwise to commit themselves to the defense of some inalienable right to sell crackpottery on YouTube, liberals would be equally unwise to focus so intently on the algorithm, and to pour so much energy into pressuring Silicon Valley to build a better, more progressive way of steering their readers through the information sea.

Whatever role the structure of the internet plays in radicalization, the root causes are still primarily sociological and political, and they will perdure and manifest themselves somewhere, somehow, no matter what YouTube suggests for your next video when you watch a Milton Friedman lecture.

Moreover, there is no better example of the technocratic spirit’s moral idiocy than the idea that, with a social crisis in the hinterland and ideological calcification at the heart of educated power, what we need is a better algorithm to prevent people — young men, especially — from getting too restless or rebellious or extreme.

The healing of the body politic will not be accomplished with such fiddling. There may be left-wing or liberal solutions to our deeper problems. But an elite that tries to manage them away with more enlightened media curation deserves to inherit nothing but the wind.

Ross Douthat has been an Opinion columnist for The Times since 2009. He is the author of several books, most recently, “To Change the Church: Pope Francis and the Future of Catholicism.”

You can follow him on Twitter: @DouthatNYT

Read More

This startup wants to sell you chic home design for less than $2K

This startup wants to sell you chic home design for less than $2K

By Jesus Diaz2 minute Read

The A-frame house is an icon. With its steep roofline, it became a wildly popular building type, particularly in vacation homes, through the mid-’50s and ’60s. Now a company called Everywhere Inc. wants to make it easier for you to build your own, selling you the plans to a neat 1,574-square-foot cabin called Ayfraym.

All the blueprints and material specifications needed to put it together are included in a cardboard box. They come both on a flash drive in PDF format and printed on 36-by-24-inch sheets. The company says that the package also includes reading materials about the house, discount codes to build some of the stuff you need at providers all around the United States, as well as a door knob, a hammer, and a hat with the company logo.

The cost? $1,950.

[Image: courtesy Everywhere Inc.]

Traditionally, if you wanted to build a cabin such as this one, you would have to hire architects and spend tens of thousands of dollars just to plan the project–and that’s just the financial outlay. Then there’s the mental toll of attempting to communicate your wishes to a professional and having them materialized in exactly the way you imagine them. The Ayfraym plans skip that step: if you like the house you are seeing in these renderings, you just buy the plan for a couple grand and go straight into the construction process.

[Photo: courtesy Everywhere Inc.]

The company claims that the total estimated construction cost ranges from $252,000 to $277,000, which includes permits, basic site preparation work, foundation work, and everything else Everywhere Inc. would need to finish the structure. According to company CEO Brand Winnie, who cofounded the company with his wife after leaving his tech job in Silicon Valley, the company has a network of partners for construction and materials through 16 states. You can look at how much it would cost in each of these states on the map on Everywhere’s page.

However, you don’t have to work with them at all if you don’t want to. You can buy the plans, and then hand them off to the contractor of your choosing or build it yourself. (The estimates above don’t include the land, so you have to add that to the total cost.)

[Image: courtesy Everywhere Inc.]

Everywhere Inc. started as a company that sold designer trailer homes, but that venture stalled. While many people are hesitant to invest money in trailer homes, they are comfortable buying standardized plans to build a permanent home, Winnie says. He hopes that blueprints-in-a-box–plus the company’s services to turn the plans into the final product with the collaboration of contractors–are a surer bet.

The inspiration for Ayfraym came from his family history: “Growing up as a kid, I would always travel to our cabin that my grandpa Fred built in the ’70s,” Winnie says. “It’s up near the Tetons on the border of Idaho and Wyoming, and it’s one of my favorite places on Earth. The cabin my grandfather built was the base camp for all things adventure in the region, and it’s created some of my best memories.”

From the renderings, it looks like Winnie and his partners have tried to capture that spirit. Now it remains to be seen whether the final product fulfills its promise of good house design at an affordable price.

Read More

Vintage Monoscope Tubes Generate Classic TV Test Patterns Once Again

Vintage Monoscope Tubes Generate Classic TV Test Patterns Once Again

Night creatures and insomniacs of a bygone era may fondly recall a TV test pattern appearing once [Jack Parr] or [Steve Allen] had had their say and the local TV station’s regular broadcast day had concluded. It was affectionately known as the Indian Head test pattern, for the stylized Native American, resplendent in a feathered headdress, that featured prominently in the graphic.

Unknown to most viewers was exactly how that test pattern and others like it were generated. But thanks to [Rich “The Lab Guy” Diehl] and his monoscope restoration project, we can all share in the retro details. It turns out that while some test patterns were merely a studio camera trained on a printed card, most were generated by a special tube called a monoscope. It functioned in basically the same manner as a studio camera, but rather than scanning the incident light of a scene with an electron beam, the image was permanently etched into a thin aluminum plate. [Rich] laid hands on two vintage monoscope tubes, one containing the Indian Head test pattern, and set about building a device to use them. “The Chief” can hold either tube in a Faraday cage of thin, flexible PCB material and 3D-printed parts, with supporting electronics like the power supply and video amplifiers in an aluminum chassis below.

It’s a nice piece of work and a great lesson in how it used to be done, and the lithophane of the Indian head is a nice touch. Hats off to [The Lab Guy] for build quality and great documentation, including a detailed video series that starts with the video below. If you need a little more background on how video came to be, [Philo Farnsworth]’s story is a good place to start.

[via @TubeTimeUS]

Read More

a quarry and a warehouse designed by esrawe studio showcase material library in guadalajara

a quarry and a warehouse designed by esrawe studio showcase material library in guadalajara

driven by the captivating nature of the project’s surrounding landscape, mexican architects esrawe studio conceived the showroom for grupo arca. the architects designed a sculptural building for an elegant retail experience in guadalajara, mexico that highlights the project’s raw materials and the company’s attractive stones. the designers also integrated technology into the project to capture visitors’ behavior and their interest in the different materials.  

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by genevieve lutkin

the architects created an access through a small opening in the monolithic facade that links the visitors with the monumental central space. the ‘arrival to the quarry’, or the forum, is the first space visitors encounter when entering the building. this area is the starting point for the buildings’s multiple exhibition halls, that showcase varying materials through different characteristic rooms. the halls articulate around the forum, creating a dynamic experience through all the spaces.

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by césar béjar / production: revista container

the proposal dismisses traditional retail experiences and instead promotes learning and knowledge, specifically aiming to understand why and how architecture, design, art and culture are created and interpreted in mexico and worldwide. the project was created as a platform that stimulates dialogue and dissemination, and establishes a relationship of mutual benefit with the community, as well as with grupo arca’s clients and partners. the building’s main objective is to promote involvement in the construction of the cultural and creative expression of mexico.

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by jaime navarro

the project was driven by two contrasting concepts – the quarry and the warehouse. the concepts are represented by two separate volumes that vary physically and functionally, yet are woven into each other to create the entire building. the quarry houses the forum, the design center, the cafeteria and the multipurpose room. the warehouse is a translucent, neutral space that acts as a canvas to showcase the stones and is framed in the background by a forest. the warehouse emphasizes the vast material product library that grupo arca owns and its meaningful connection to nature.

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by jaime navarro

technology plays an essential role in the proposal because it allows the interaction of visitors with materials through QR codes. once the viewer scans the code of the desired material, it provides the user with the information, description, and the costs of the selected materials. this creates a database of the materials of highest interest for the visitors as well as each of the client’s viewing history enabling the opportunity to analyze trends and purchasing behavior.

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by césar béjar / production: revista container

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by genevieve lutkin

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by jaime navarro

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by jaime navarro

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by genevieve lutkin

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by césar béjar / production: revista container

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by genevieve lutkin

esrawe-architecture-studio-grupo-arca-museum-guadalajara-mexico-05-16-2019-designboom

image by genevieve lutkin

project info:

creative director: héctor esrawe

architectural concept: esrawe studio architecture

interior design and furniture: esrawe studio

design team: laura vela, antonio chávez, maría santibañez, jacobo mendoza, andrea guillén, brenda vázquez, fernando carnalla, tatiana godoy, fabián dávila, daniela pulido, javier garcía rivera

showroom experience concept: esrawe + cadena

brand identity: cadena+asociados concept design

renders: yair ugarte, moisés gonzález, gael félix

technical collaboration: a-001

external advisors: culdesac, alberto martínez

lighting design: luz en arquitectura

landscape design: paar paisaje arquitectura

construction: cdm, jaime de obeso, olmo ernesto godínez

photography: genevieve lutkin, jaime navarro, césar béjar (production: revista container)

designboom has received this project from our ‘DIY submissions‘ feature, where we welcome our readers to submit their own work for publication. see more project submissions from our readers here.

edited by: cristina gomez | designboom

Read More