There's the America that voted for Hillary, and America that voted for Trump. Now, the New York Times has mapped—and named—both.
If there's anything the 2016 Presidential Election taught us, it's that our country is far more divided than anyone had really guessed. A new visualization from the New York Times makes that divide starkly clear, separating the districts that voted for Hillary and Trump into their own distinct geographic realities. If you're a Democrat, the results are bleak: Hillary supporters are an archipelago drowning in the great American ocean of Trump.
Hillary's America, which is made up of almost all of America's cities, roughly looks like the Philippines or another Southern Pacific island group. Geographically, the largest almost-continuous areas of Clinton voters run along the coasts, broken up by swathes of Republican-voting districts, which The New York Times has given place names inspired by nautical markers—such as the Wyoming Shallows, the High Plains Sea, Bakersfield Bay, or the Alabama Gulf.
Trump's America, on the other hand, makes up for about 80% of the country's landmass (although a minority of its population). It almost looks like a regular map of America, except the coasts are worn away, and there are several large inner lakes where Hillary voters congregated. It is, perhaps, no coincidence that Trump's America looks similar to what America itself will look like if global sea levels continue to rise, drowning out our nation's coastal cities. The part of America most likely to be visibly affected by climate change voted strongly for Hillary, while in-land America voted for a man who vowed to dismantle the EPA and withdraw the United States from the Paris Agreement.
Such a map is depressing, true, but one thing to keep in mind is that the political breakdown of America isn't as startlingly red and blue as maps like this make it appear. In most American voting districts, a healthy balance between Republicans and Democrats is the norm, not the exception. Trumpland may be big, but it's not really as big as this mapped perspective seems. And it's sure as hell got a lot less people than Hillary's real America.
Rejoinders puts the old Dutch masters' most grotesque portraits in iMessage.
It's hard not to love an emoji app with a change log like this: "Updated to incorporate a sobbing toddler emoji painted in 1670 by Jan Steen, graciously furnished by The Mauritshuis in The Hague." But that's the awesomeness of the Rejoinders in a nutshell. It's a $0.99 iMessage sticker pack that repurposes the bucolic, dour, phlegmatic, or downright bilious visages of people painted during the Dutch Renaissance as hilarious emojis perfect for use in emails and text messages.
Created by Molly Young, a copy director at Warby Parker, and Teddy Blanks of Brooklyn-based design studio Chips, Rejoinders is made up of 40 faces, plucked from paintings by old Dutch masters like Bruegel, Steen, Vermeer, and more. For the most part, the emoji seem like they'd be most at home on the perturbed end of the emoji spectrum, in place of old favorites like the grimacing or neutral smiley. However, there are Rejoinder stickers appropriate for more joyful moments, including the Mephistophelean visage of a laughing baby, face utterly deformed by malevolence.
Speaking to Quartz, Young says it was this baby's face which actually inspired her to make Rejoinders. She discovered it in a painting on auction at Sotheby's, which she eventually was able to buy. "I totally understand this man, with his mischievous grin," she says. "His rotting teeth." Soon after, Quartz explains, she was using the rotten-toothed visage in office communication. What began as an instinct to troll soon became one of the App Store's best-selling iMessage apps.
In 1989, Apple hired Thomas Rickner to do something revolutionary: put print-quality typefaces on every computer.
When Thomas Rickner decided he wanted to be a type designer after a lecture on the history of Baskerville at the Rochester Institute of Technology in the mid-1980s, his typography professor had some simple advice: Don't do it.
"He told me it was a road to frustration," recalls Rickner, who nearly three decades later is a font production manager at Monotype. "At the time, there was just no easy path to being a type designer. You couldn't go to school for it, and there were only a handful of companies around the world that would teach you. He told me to become a book designer. That, he said, I could do anywhere."
Clearly, Rickner ignored his professor's advice. And it was a good thing he did. Over the next two decades, thanks to the advent of personal computers, type design—and, consequently, the ability to become a type designer—exploded. Now, not only do dozens of universities offer type design degrees, but it's possible to freelance as a type designer, selling your fonts online, without ever being hired by a major foundry.
What changed? There are dozens of reasons why the last 30 years have seen a renaissance in type design. But when you talk to Rickner, you get the sense that one company played an outsized role in democratizing type during those years: Apple. And he had a part in it.
Rickner's first type design job was at Imagen, a laser printer manufacturer, where he started in 1988. At the time, laser printers didn't just print any font your computer had installed; instead, printers all had their own built-in fonts, custom designed for each individual model. Rickner's job was to "hint" these typefaces, or write little programs that optimized the way characters were displayed at different font sizes on Imagen's laser printers.
His experience with hinting led Rickner to a job at Apple, where he became lead typographer in 1989. It was an important time to be a type designer at Apple, because the Mac maker was about to do as much to revolutionize computer typography as it had done to revolutionize desktop publishing. Apple was aiming—in secret—to build a way to render third-party fonts right into the Mac operating system itself, radically changing the way computers dealt with text.
Let's back up a second. When Apple first released the Macintosh in 1984, one of the many ways in which it was revolutionary was in allowing users to choose between fonts. These fonts were extremely limited, though. Until 1991, the Macintosh only supported bitmap fonts—or fonts made up of individual images, one per character. When you made a bitmap font larger or smaller, it became harder to read, either because it had become pixellated at large sizes, or because the operating system just shrank all the pixels down without doing any optical correction to make sure the letters were still legible. The Mac only natively supported a small number of low-res proprietary fonts. These were just good enough for someone printing out their grandma's cookie recipes, or writing a family newsletter, but not for more serious type—let alone professional—aspirations.
When Rickner was hired by Apple in 1989, though, he was hired for a secret project: TrueType, an attempt by Apple to radically amp up MacOS's font rendering abilities. Just like the laser printers Rickner had previously worked on, TrueType fonts didn't use bitmaps, but instead, were based on outlines. In other words, when the computer displayed a font on screen, it didn't just show you a selection of low-res image files. It actually drew the font on the screen, according to a mathematical formula dictating what it should look like. That meant TrueType fonts could scale to any size or resolution. Even better, they could be hinted, making sophisticated typography possible on off-the-shelf computers for the first time. It also opened the door for all sorts of classic fonts that had been used in print for years (like the first three TrueType fonts: Times Roman, Helvetica, and Courier) to finally make their digital transitions.
Apple debuted TrueType in 1991, and to make sure it became a ubiquitous standard, it even licensed it to Microsoft, which introduced TrueType fonts with Windows 3.1. Quite suddenly, it became possible for millions of computer owners to display and use rich, sophisticated, and scalable typefaces created by the world's greatest type designers and type foundries. This—along with the rise of user-friendly font design programs like Fontographer, which made creating professional quality digital typefaces easy—was more responsible than anything else, says Rickner, in "the democratization of type design."
"It's hard to overstate how revolutionary this was," says Rickner. Before, if you wanted to use quality typefaces on a computer, you needed to buy expensive professional software from a company like Adobe. It was something only graphic design professionals ever bothered with. "With TrueType, Apple said: No, type should be part of the body of a computer, not just the clothes it wears." In other words, type rendering needed to be a core part of every operating system, as integral to the machine as copying files, drawing graphics, or managing memory. Now, all of a sudden, the average computer user didn't just have access to 8 or 10 low-res fonts. They had access to hundreds, every bit as good as the ones they saw in magazines and books.
Rickner left Apple shortly after TrueType shipped, and joined Monotype in 1994, where he has worked ever since. He was heavily involved in the recent creation of variable fonts—a single font that acts like multiple fonts, depending on the circumstances, which has the backing of Adobe, Google, Microsoft, and Cupertino. Ironically, this new standard is an extension of work he was doing at Apple when he left the company in 1992.
Today, remembering his professor's caution that he'd never make it in this business, he says he's surprised to find himself the oldest guy in a room full of Monotype's young type designers, most of whom fell in love with type—and are capable of pursuing it as a career—because, in part, of a project he helped Apple with over a quarter-century ago. "I decided I wanted to be a type designer after attending a lecture on the history of Baskerville, and being blown away by the revelation that something so utilitarian that I took for granted every day could have been designed by someone," Rickner says. Thanks to TrueType, it's a revelation that has now been shared with a new generation of type designers.
You probably know that the websites you visit—even this one—track your behaviors in some capacity. Maybe you think it knows your browser, your computer type, your connection speed, maybe even your computer type. But webpages can track far more than that. In fact, they can analyze your behavior so well that even with no other information at all, they can tell it's you visiting a site, not anyone else.
Clickclickclick.click is a sardonically narrated website that transparently exposes all the ways in which a website can track you based upon your behavior, created by Luna Maurer & Roel Wouters of Studio Moniker. The user-facing site itself basically just a big green button and a terminal telling you what the server is "seeing" as you browse the site. But it turns out that the server can see almost everything: How you move your mouse, what you click on, how quickly you click on it, when you browse away, when you come back, and many more things besides.
But the way Clickclickclick.click can extrapolate these seemingly mundane details into nearly psychically accurate user profiles is what really amazes... and horrifies. For example, based upon the way she moved her mouse, Clickclickclick.click was able to guess my colleague's gender accurately. It can also predict things like age range, computer literacy, probable income and educational level, and more, just by the way you browse the site. When Clickclickclick.click draws a new conclusion about you, it ruminates on it aloud in an amusing German voice.
Are these conclusion 100% accurate? No, but they're strong guesses informed by the behaviors of billions of users across trillions of sessions. In fact, by building a profile of a user's unique web-browsing tics, it's possible for entities like advertising networks to track users across multiple websites—for example, to determine that you're a woman living in New York in your late fifties who likes puppies and is looking for a new car. And from there, it can serve you up the perfect ad.
Sound implausible? It isn't. At least in part, this is how companies like Amazon and Facebook seem to offer you up the perfectly relevant product or ad at any given time, even when you've never searched their sites for what they're offering. So if it scares you that Clickclickclick.click can correctly guess so much about you, how much more terrifying is it when sites that already know so much more about you start tracking you this way?
According to Wouters—who also gave Clickclickclick.click its delightfully wry voice—raising awareness of the way companies can track people across the web was the reason Studio Moniker created the site in the first place. "We are fascinated by the ease with which people allow a few big companies to monetize our digital behavior, as well as the endless possibilities that creates," he says. "We hope that our visitors start thinking about that as well."
Audio design company AIAIAI believes modularity has a place in electronics. Here's why.
When Apple unveiled the iPhone 7 back in September, it also forced the obsolescence of a 50+ year standard: The 3.5mm audio jack. The sound you heard—beyond the blogger complaints—was that of a million pairs of corded headphones being flung to the back of the drawer, suddenly incapable of natively working with all future models of the most popular brand of audio players on Earth.
But if you had a pair of AIAIAI's modular TMA-2 headphones, which were specifically designed so that the cables, headband, speakers, and earpieces could all be upgraded over time, your smartphone dropping support for a connector didn't matter, because you knew it would only be a matter of time before the TMA-2s would support the iPhone 7, headphone jack or no. And sure enough, just two months later, AIAIAI has announced the H05, a replacement headband for the TMA-2 that gives the headphones wireless Bluetooth audio support.
A small Danish audio company founded in 2006, the TMA-2 is AIAIAI's first modular product—a response to the long list of tweaks customers asked after it released its first headphones, which were not upgradeable, in 2010. "When we released the TMA-1 we got a lot of feedback," says AIAIAI CEO Fredrik Jørgensen. "We were given lots of new ideas, as well as requests for small adjustments from our network and users. But with a conventional design, it's very complex to do a mark 2, let alone a large portfolio of different product versions. That's why we turned to the modular approach."
The way Apple suddenly dropped support of the 3.5mm jack—a probable sign that the rest of the industry will soon follow—shows the wisdom of AIAIAI's modular design thinking. By being able to adapt its current, on-the-market product without resetting the manufacturing process, AIAIAI will be able to weather an industry-wide sea change in a way that might be impossible for other smaller headphone manufacturers. "Modularity gives us agility, and helps us avoid ending up with a lot of obsolete inventory when the market changes," says Jørgensen.
But modular design is also more consumer-friendly. When a pair of TMA-2s has a loose connector, or one of the speakers blows out, customers aren't left facing a total replacement; they can easily replace the broken module themselves. Likewise, modularity encourages a broader customer base: The TMA-2s can be easily customized in configurations ranging from sub-$150 cans perfect for Beats lovers who want to step up their audio game, all the way to $250+ monitors aimed at pros. That's without taking aesthetic customization into account. As long as you like the general design language of the TMA-2s, they can be easily modified to match any individual's personal style. This, in turn, gives customers confidence that if they do invest in a pair of TMA-2s, their money won't go to waste.
Not all hardware products are suitable for a modular approach, though. Designs that need to be as thin and light as possible—for example, smartphones—are much less practical to make modular, for all sorts of reasons. Google put tens of millions of dollars into designing a modular smartphone called Ara, yet killed the project earlier this year, for unstated reasons—reasons Lars Larsen, founder of Kilo and designer of the TMA-2s, think has to do with complexity.
"Even though it sounds so simple, trying to account for as many parameters [as a smartphone] means that modularity can become very complex, very quickly," Larsen says. "For us, in this product category and the size of the company that we are currently looking at, it works, because it's still pretty basic and controllable." In other words, AIAIAI isn't trying to build a slim, light, modular computer that can be anything for anyone. Its goal is to make a modular gadget that does only one thing well, in a market known for its tolerance of chunky designs. (And even so, it's obvious why AIAIAI is only trying to design modular headphones, not modular earbuds.)
But though designing more complicated modular hardware like Google's Project Ara can be like lopping the head off a hydra, Jørgensen thinks big companies like Google and Apple will still eventually need to learn how to do it as well as smaller companies like AIAIAI. "The whole world can't throw out millions of fully functional smartphones every year, when 80% of the newest model is functionally and even materially identical to the last," he says. "Our planet simply can't withstand the current level of waste."
The rooms were to designed to look as if they just sprang from the pages of Patricia Urquiola's own notebook.
In part due to its annual design fair, designers have long had a love affair with the Italian metropolis of Milan. Now, Patricia Urquiola has turned the interior of the Room Mate Hotel into a designer's love letter to its host city, replete with pink marble flooring that matches the nearby Milan Duomo, and walls that draw inspiration from the terracotta and terrazzo of the local architecture.
Occupying a renovated building from the late 19th century across the street from one of the world's oldest shopping malls, Galleria Vittorio Emanuele II, Urquiola designed the Giulia Hotel Room Mate's interiors as an ode to Milan's rich pedigree in art, architecture, and design. Many of these nods are done through patterns: for example, a crisscross pattern near the elevators that was done to remind visitors of Sforza Castle's iconic brick towers. Other references are even more abstract: according to Frame, a dashed grid motif running through the hotel is Urquiola's way of paying tribute to Milan's long history in the graphic arts.
Each of the Room Mate Hotel's 85 rooms has been designed to make visitors feel as if they are staying in a genuine Milanese home. Each room is filled with bespoke, handmade furniture from Italian furniture brand Cassina, and the walls are covered in exclusively Milanese artworks. Throughout, the rooms use a mixture of the dashed grid and muted paint colors, again evoking the natural color palette of Milan. The walls and ceilings, in other words, are made to look as if the room itself sprung to life from a designer's dotted grid paper.
Urquiola herself, of course, is no stranger to Milan's charms: she's had her studio in the city since 2001. That makes her the perfect person to design Milan's newest design-friendly hotel. The Italians are famous for their hospitality, but thanks to Urquiola's work, now you can stay in a genuine-feeling Italian home even when you don't know anyone in Milan. And the host of that home has some pretty exquisite taste. You can book a reservation here.
To laymen, architectural principles are rather abstract. Buildings seem to stay up based upon nothing but sturdiness, when in actuality, even the simplest buildings stay erect based upon a combination of structural principles, many of which are extremely hard to explain in words—the way that things like beams, columns, trusses, retaining walls, and more actually work.
What might surprise you is that these structural principles don't seem abstract just to laymen. Many an architecture student wrestles with them too, which is why Brazilian architect Márcio Sequeira de Oliveira created the Mola Structural Kit a few years back in response to the abstract nature of the postgraduate course he was overseeing. Now, this set of Tinkertoys for architects is back, as ArchDaily points out, with an updated kit that is even more realistic.
Like the first kit, the Mola 2 is made primarily of a series of magnetized springs, planes, and ball bearings which, when assembled on a base, allow architects to visualize the way structural forces work on a building—and how they deform. Two new mechanisms in the set join the previous collection of pieces to allow architects to simulate continuous pillars, continuous beams, transition beams, and more.
The Mola 2 is fully backwards compatible with the first kit, and can be used to mock up some cutting edge buildings, such as OMA's CCTV Headquarters in Beijing, or Arata Isozaki's Art Tower Mito in Japan. It can also be used to model the structural forces that act upon infrastructure like cable-stayed bridges; press on your bridge with a finger to simulate how a catastrophic tsunami or hurricane might break it apart.
The Mola 2 Structural Kit can be preordered on the Brazilian crowdsourcing site here, starting at around $36.
Design agencies are globalists. The Western world is increasingly isolationist. How will that tension play out?
Donald Trump, who will become the 45th president of the United States on January 20, seems like a distinctly American phenomenon. Yet he's not.
From the U.K.'s vote to leave the European Union earlier this year, to elections in France and Austria this year where votes to keep far-right candidates from becoming heads of state ride the razor's edge, the world is dealing with a wave of isolationist kickback after almost two decades of globalist expansion.
It's a wave the design community, which is traditionally globalist, is going to have to weather. In America, this means that design firms are facing the prospect that their workforces, a huge percentage of which are made up of foreign-born talent, might face new scrutiny under the Trump administration. It also means that firms' values will be tested like never before.
So far, Trump hasn't laid out many concrete plans, which is causing an air of uncertainty in the design world. "We don't know yet what President-elect Trump plans to do. We haven't seen any specific proposals from him yet," says John Barratt, president and CEO at Teague, where 15% of the firm's 300-strong workforce is foreign-born. It's still mostly just rhetoric. But there are some guesses that can be made.
Trump's campaign promises for his first 100 days in office only specifically mention repealing Obama's executive order on Deferred Action for Childhood Arrivals, which could see between 740,000 and 1.7 million people who came illegally to the United States as children deported. He has also promised to increase scrutiny of immigrants from "terror-prone" countries. What Trump means by this and what specific countries would be included in this list is anyone's guess, but obvious contenders include Iraq, Afghanistan, Pakistan, Somalia, Indonesia, and other terrorist safe havens.
Neither of these plans is likely to significantly impact most American design agencies. But there's general anxiety among design agencies we spoke to that Trump's administration could make applying for H-1B nonimmigrant work visas more difficult, something Trump has repeatedly promised before and after the election.
These visas are the most common method of bringing creative professionals into the country to work legally, and design agencies depend upon them extensively. This anxiety is nothing new, but it's deepening in the face of a Trump presidency.
Craig Dykers, founding partner and executive director of Snøhetta—where 20% of its U.S. workforce could be affected by more stringent immigration controls—points out that scrutiny of H-1B applicants has been getting tighter and tighter every year since 9/11, including under Obama. In 2001, 161,643 H-1B visas were approved by the State Department; 13 years later, the number of available visas had grown by just 3%, compared to a U.S. population growth of about 10%. So it's already pretty hard to get in. "The big question is: Will it get more stringent?" says Dykers. "And we don't know how to answer that yet, except to say, it's already plenty stringent."
If H-1B and similar visas become harder to get, design firms will likely find it more difficult to fill their ranks with top-quality talent. At Siegel+Gale, 10% of all U.S. employees are born abroad. These employees, says co-CEO and chief creative officer Howard Belk, are "impact players"—designers who count among Siegel+Gale's VIPs. "We simply would not bother going through all the hassle of trying to bring them to this country if they weren't worth the effort," Belk says, pointing out that every H1-B worker comes with an overhead in legal fees and paperwork. So even if only a small number of foreign-born professionals are denied entry or reentry into the country, a tougher policy could have an outsized impact on an agency.
[Photo: skynesher/Getty Images]
This raises a larger point. Like a particle accelerator, design only reliably creates new elements when it is blasting as many different atoms, molecules, and quarks off of each other as possible, except those particles are different creeds, cultures, and heritages. "Design agencies especially benefit from having an international staff," says Dan Harden, CEO of Whipsaw, where a full 40% of the team is foreign-born. "When you mix different cultures in a design process, you get creative alchemy. You'll be challenged, surprised, and forced out of your own box, but you'll innovate."
So design is a global industry. Meanwhile, the platform that got Trump elected is isolationist at best (and, many would say, white nationalist at worst). Outside of the mere disruption of bringing foreign-born talent into the United States, there's an inherent conflict between these worldviews, which has both a downside and a surprising upside.
The downside? "On at least some level, the biggest concern the design community should have about a Trump presidency, and other isolationist movements like Brexit, is that it creates a perception that these nations are less open, inclusive, and diverse," says Ije Nwokorie, CEO of Wolff Olins, where 33% of the U.S. workers are foreign nationals. "That's kryptonite for the creative process, because creative people like to work in diverse environments. That's where they feel safe. A feeling of diversity is why London, New York, and San Francisco are such creative hubs." So even before you get into the hard issues of immigration policy, isolationist rhetoric like Trump's can make designers more fearful, and less likely to pursue jobs in America. Pentagram's Natasha Jen puts it bluntly: "What's worrisome is the anti-immigrant, if not white supremacist ideology that will make America a much less attractive, if not a scary, place to work and live."
But there's a surprising upside too, according to Nwokorie. "Just like in art and other forms of creativity, design often thrives when it has something to oppose, to fight for or against," he says. "Benign, unchallenging environments can also be kryptonite for the creative process. So if there's a silver lining in Trump's election and Brexit, it's this. At least in some degree, the design community is interpreting these events as an attack on what we stand for. I can already feel an awakening after the initial shock in the U.S. and the U.K. to double down on our efforts to create things for the many, not the few."
This feeling of having been called to action—creatively and otherwise—is one echoed by many design firms. At Snøhetta, employees are being openly encouraged to attend protests, and even having their travel fares to upcoming marches in Washington, D.C., subsidized. And after pointing out that some politically motivated companies like Patagonia even post bail for employees (provided they first complete complementary Ruckus Society training on the correct ways to protest nonviolently), Dykers said: "I wasn't thinking about that before, but I sure as hell am now."
Regardless, while the future may be uncertain, none of the design agencies we spoke to had any intention of changing the way they did business. Some were openly defiant about it, like Christof Mees, CEO of Ziba, where at least 25% of its employees working in America are foreign-born. A German with a green card himself, Mees is shocked by Trump's rhetoric against immigrants and foreigners on multiple levels. "We will not change our principles," Mees says. " We will not turn into an all-white male firm. We will double down on our efforts. We need to drive diversity, because it's right for Ziba, and it's right for our society."
Even if design firms wanted to change how they did business to cater to Trump's America, it's not entirely clear that they could. Design can't be done in a bubble; internationalism is just too intertwined in the way design firms do business, and the very process of design. Just look at some of the most prominent designers and designs of this century. Architects, like Bjarke Ingels and Norman Foster, routinely work outside of their home countries, and employ incredibly diverse staffs. Jonathan Ive, Apple's design chief and the closest thing design has to a household name, is British. Located in Washington, D.C., the new National Museum of African American History and Culture was designed by David Ajaye, a Ghanian-British architect, in collaboration with Philadelphia-born Philip Freelon.
If a Trump administration stops these sort of creative synergies from happening, the long view is that designers will simply go do their work in new places, where diversity is celebrated, not strangled. This, in turn, could lead to the rise of new creative hubs around the world—just not in the U.S.
No matter what, the firms we spoke with were adamant that nothing about Trump will ultimately change design's globalist ethos. As Wolff Olins' CEO Ije Nwokorie explained to me: "Trump the man is just too small a notion to change what we, as designers, stand for."
Inspired by Italo Calvino, a team of engineers teaches neural networks to imagine entirely new cities.
From above, each city has its own distinct character: something beyond the sum of its parts, a unique fingerprint of streets and houses that makes it quintessentially itself. Even if you didn't know the landmarks, you'd be unlikely to mistake a satellite image of, say, Milan, for one of New York or Hong Kong. They just have different styles.
Now imagine if an aerial image of New York could camouflage itself as one from Hong Kong, or Milan. That's just one of the things Invisible Cities, a new project from Italy's Opendot (which includes Gene Kogan, Gabriell Gambotto, Ambika Jib Samsen, Michele Ferretti, Andrej Boleslavsky, Damiano Gui, and Fabian Frei), can do. It's a neural network that can not only transfer the style of one city onto a map of another city . . . it can dream up realistic satellite imagery of entirely new, imaginary cities.
The backbone of Invisible Cities is pix2pix, machine learning code that radically changes the style of an image—making a sketch of a handbag into a realistic photo of a handbag, for example, or translating a picture taken by day into one taken at night. In Invisible Cities' case, though, the bot has been trained on a MapBox database of aerial satellite images of cities, along with their corresponding map tiles of OpenStreetMapData, which color-code the images by roads, parks, buildings, and more.
//invisible_cities_patchwork_02
By learning the relationship between the satellite images and the map tiles, the bot is able to take a map tile from one city and give it the "style" of another. For example, give a more Mediterranean city like Naples the industrial look of New York. Even more interestingly, Invisible Cities can generate entirely new realistic-looking satellite images based on hand-drawn sketches. In other words, creating an imaginary city is as easy as sketching down some lines on a piece of paper, then feeding it into the bot, which anyone can do through the publicly available code (technical knowledge allowing).
Given the name of the project, it's no surprise that the team was inspired by Italo Calvino's 1972 novel Invisible Cities, which is largely made up of 55 prose poems about mythical cities, which are in turn inspired by the fanciful (and mostly made-up) cities that Marco Polo allegedly visited on his trip across Asia in the 13th century. "We thought it was an appropriate analogy as we started making these generative city maps," Kogan says.
Since the technology can be used to create realistic fictional cityscapes with minimal effort, it could be useful for video game developers or even Hollywood special effects artists. Shorter term, though,the team says they might bring Invisible Cities to the web, making it easier for people with less technical acumen to bring the fanciful metropolises of their minds to life.
In the new book Let's Play, artistic director Christiane Nill and photographer Lionel Henriod task 100 Swiss creatives who work in three-dimensional fields like architecture, sculpture, and industrial design to play with blocks for the camera then document what they make.
The parameters of Let's Play are simple. Designers are given 270 simple wooden blocks which have been custom created for the project. All blocks are visible at the beginning of each play section, and they're organized into groups of 34 different shapes. Each player has 30 minutes to construct something using any of the blocks they want, plus three picked by the previous player—those must be incorporated into the structure. (The first player used blocks selected by the carpenter who made them.) Beyond that, anything goes.
According to Nill, who was a founder of Lausanne's mc² studio back in 2009, the idea for Let's Play came after a session spent playing with her daughter using the blocks she grew up with as a child. She was at her mother's home for the holidays at the time, and as she and her daughter played, one adult after another—her husband, her mother, her sister, her uncle, her aunt—came by to play with the blocks themselves. Taking note of this phenomenon, Nill further noticed that "the more I observed the details of their constructions, the more I saw they had an astonishing resemblance to their creators." It was only natural from there she'd want to see what designers did in the same circumstance.
Filip Leu
When it comes to playing with blocks, there are no universals, but some trends come to light. Architects tend to make playful buildings based upon advanced structural principles that would be extremely costly to make in real life. Political artists make orderly constructions, almost like statements. Industrial designers can get extremely wacky and topsy turvy with their structures, creating buildings that would be seemingly impossible to reproduce.
In short, one theme of playing with blocks seems to work against whatever constraints apply to designers when they work within their field. Which makes sense. When you play, you want to play, not work. And that's as true in design as it is anywhere else.
You can order a copy of Let's Play online for around $65 here.
The Tangible Media Group envisions a future where building any device is as simple as tying two strings together.
Last year, we hailed LineFORM—MIT's bizarre, snake-like robot—as the possible future of UI. The idea behind the device was deceptively simple: What if the integral building block of computing wasn't a rectangular block, like your smartphone, your tablet, or your laptop, but a robotic string of blocks that could transform its shape depending upon need? It could take on the basic form of a phone when you need to make a call, or morph into a wrist-worn fitness tracker when you're exercising, or any number of other form factors.
Now, the LineFORM has evolved. Meet the ChainFORM, a modular version of the original which also comes equipped with integrated sensors and a low-res display. The ChainFORM could also be infinitely expandable, in theory: You could add as many modular links as you want, transforming the chain on the fly from the most adaptive computer accessory ever to a flashy wearable exosuit that potentially augments your strength. The updates take it from a mechanical prototype to something more functional, resembling a usable gadget.
ChainFORM was created by Ken Nakagaki of MIT's Tangible Media Group, the previous designer of the LineFORM, alongside SensorTape creator Artem Dementyev of the Responsive Environments Group. Each ChainFORM module is a small robot link with actuators on each side, allowing it to twist and bend at any angle when attached to other modules. To grow the chain, you just snap on more modules. Right now, ChainFORM only supports 33 links, but that's just a limitation of the prototype based upon power draw. In theory—say, by giving each battery chain its own small battery pack—a ChainFORM chain could be made up of any number of individual links, each one of which also contains a number of LEDs and sensors, detecting things like pressure, touch, rotation, and more.
How do Nakagaki and Dementyev envision this expandable, adaptable technology actually being used? Right now, they see two major applications for the ChainFORM.
One is as an infinitely adaptable and extendable computer accessory. In this mode, depending upon the application, the ChainFORM could transform from a mouse-like control puck, to a keyboard-like input device, to a VoIP phone handset, to a second display. Think of the new MacBook Pro's TouchBar as a three-dimensional robot, and you've got a decent idea of what MIT's imagining here. The ChainFORM is just a prototype, and clunky in some ways, but it's still a fully-operable iteration of that vision, only in need of some finesse.
Like the LineFORM before it, Nakagaki and Dementyev imagine their ChainFORM being used as a prototyping tool. More interesting, though, is the way they envision it as functionizing as a customizable robot skeleton which could be used to animate and control analog objects. For example, you could stuff an old teddy bear with pieces of a ChainFORM to create your very own Teddy Ruxbin, or attach them to a camera to make a walking surveillance bot.
The duo—seemingly inspired by Starship Troopers—have yet another idea on how the ChainFORM could be used: As wearable, modular power armor. This was an idea Nakagaki explored with the LineFORM, which could be used as a workout device by wrapping around the arm and providing resistance. The ChainFORM could take this notion to the next level, wrapping around a person's entire body and using its robot actuators to help build or augment strength. You can snap off links from the ChainFORM and attach them to your glove to make a customizable exoskeleton, or take a long link and tape it to your spine as a posture connector.
The ChainFORM isn't there yet—like most of the Tangible Media Group's experiments, it's a prototype meant as an experiment with future interfaces, not a consumer-facing product. But Nakagaki and Dementyev both think they've created a viable vision for the future of computing and robotics: One in which building any device you want is as simple as tying two strings together.
This is what the greatest economic divide since the Great Depression looks like.
Donald Trump, soon to be the 45th president of the United States, lost the popular vote by 2.5 million votes. He lost the vote in the nation's largest 100 counties by 12.6 million votes. He overwhelmingly lost the vote among college graduates country-wide. He lost the black, Latino, and Asian-American votes. Ninety percent of the people in his own city voted against him. Only six publications in the entire country were willing to endorse him.
Add this data point to those sobering statistics about just how lopsided Trump's win was: According to research done by Mark Muro of the Brookings Institute, a nonprofit public policy organization, Trump also lost the economic vote: 65% of America's GDP voted against the self-proclaimed billionaire.
To put this in perspective, this has literally never happened before. It is simply unprecedented, says Munro, for a losing presidential candidate to have represented so large a share of nation's economic base. If votes were real American dollars, Clinton's allotment of the GDP would allow her to outspend Trump two to one. To put it in perspective, in 2000, Al Gore—another popular vote winner—won only 54% of the country's GDP, despite winning 1,000 more counties than Hilary won in 2016.
Why did Hillary carry so much of the country's GDP? As the Brookings Institute's report makes clear, it is largely because the Clinton campaign carried the vast majority of urban counties. In fact, with the exception of Phoenix; Fort Worth, Texas; and some of Long Island, New York, Clinton won every large-sized county economy in the country, leading Munro to sum up:
[T]he stark political divide underscores the likelihood of the two parties talking entirely past each other on the most important issues of economic policy. Given the election map we revealed, the Trump administration will likely feel pressure to respond most to the desires and frustrations of the nation's struggling hinterland, and discount the priorities and needs of the nation's high-output economic base.
The truth is that this country is looking at the greatest divide between urban and non-urban America since the Dust Bowl. It's easy to see echoes of today's urban/rural divide in the "dirty 30s" when New Yorkers, irritated by dust clouds blown into their city from thousands of miles away, suggested that they should send all of their old automobiles to Oklahoma to help keep the dirt down.
Like then, we need to actually reach out to help the people being left behind economically in rural America—something that will, ironically, no doubt be even more difficult with Trump in office.
Moleskine Smart Writing Set
Being productive in the 21st century means effortlessly switching between digital and analog tools (unless you're the president-elect). Moleskine's Smart Writing Set erases the divide between paper and digital by guaranteeing that everything you write in your paper notebook is seamlessly transferred to your iPad in real-time. Write an appointment in your notebook, have it automatically added to your iCal; start a sketch on paper, finish it up in Photoshop. Moleskine's got the right idea here: If paper's going to survive, it's going to do so as an interface, not as a medium exclusive to itself. Get one here for $199.
Magnetips Pens
Upgrading your pen game is worth it, but the biggest issue with spending more on pens than a Bic is how easy they are to lose. London-based design studio Typica has a solution: These magnetic pens (called Magnetips) aren't just gorgeous fine-liners in their own right, they'll stick to anything a magnet will to prevent you from losing them—a desk lamp, the side of your iPad, or even another Magnetip. $39 buys a set of 20.
Adobe Ink & Slide
If you're looking to pick up something for a digital designer, consider this $30 Ink & Slide set from iPad. For less than the third of the cost of an Apple Pencil, you get a pressure-sensitive stylus perfect for digital drawing, and what I'd argue is even cooler: a digital ruler that makes it easy to create the perfectly angled geometric lines you want.
Daily Fiction Collection
Got a lover of Scandinavian design in your office? The Daily Fiction Collection from Danish design company Normann Copenhagen offers an array of notebooks, envelopes, pens, pencils, and other office accessories that come in 18 different colors and an array of patterns and finishes ranging from mesmerizing terrazzo prints to striped fabrics. Notebooks in the collection can be purchased for as little as $14.50.
Surface Dial
Here's the rare gadget exclusively for the Microsoft crowd. The Surface Dial is a Bluetooth knob you can stick on any surface, to control any Windows 10 app or setting, just like you might control volume on a stereo. Obvious uses include scrobbling through a video, pumping up a Spotify track, or playing impromptu DJ, but design creatives can use it to up levels, switch layers, or manipulate 3D interfaces. Anyone who works with digital interfaces all day will thank you for it if you get them one. Buy one for $99 here.
Calendar Watch
Beautifully designed by Japanese art director Masashi Kawamura and Italian industrial designer Umberto Onzahe, by What? Watch is a perfect way to blend a digital calendar with a physical watch. Behind the analog dials, it puts a low-power e-ink display on the watch face so a wearer can see the outline of their day. The perfect gift for the stylish executive with enough of a design eye to realize that wearing an Apple Watch still looks pretty doofy. You can buy one for $299 here.
Philip's SceneSwitch LED Bulbs
Most of us only have one kind of light, but the sort of artificial light that's good for keeping you productive during the day can wreak hell on your circadian rhythms at night. To stay productive, you need light that matches the warmth or coolness of the sun at any given moment in the day, according to what our bodies expect. Philip's new SceneSwitch LED Bulbs only cost around $8 each when you buy them in a four-pack, and they make it simple for any lamp in the house to give off healthy, restorative light, at any time. Consider it as a stocking stuffer: Your loved one's circadian rhythm will thank you for it. Buy one here.
Designed by an Apple alum, the RokBlok is like a Roomba for record lovers.
In an ever more mobile world, vinyl—the least mobile audio format this side of a symphonic orchestra—is making a surprise comeback. But short of a return to 7"s, is there any way to make vinyl more mobile, and less dependent upon a bulky home stereo? The RokBlok might just be the answer. This boxy little robot with a needle on its belly can play records on any hard flat surface, no turntable required... although a Bluetooth speaker helps.
The RokBlok is the brainchild of Logan Riley, an Apple, Pandora, and LearnUp alum who started a company called Pink Donut a few years ago to bring his notebook of self-described "crazy ideas" to life. His company's first app, LookFor, allowed people to find their friends just by flashing colors on their smartphone, a couple years before Lyft baked that same functionality into their app to flag down drivers. His podcast, One City Under Karl, uses binaural spatial sound—or sound recorded with two microphones to simulate the way we hear—to transport listeners wearing headphones to actual physical places in San Francisco.
Riley says the idea for the RokBlok came to him during a day in the park with friends after record shopping. Riley says he wanted to share the records he'd just bought with the group, but there was no way to do so. "What was I going to do, invite half-a-dozen people back to my 300-square-foot San Francisco studio?" He went home and dreamed up a design for a little motorized speaker that could skate on the surface of a record, playing it without a turntable.
Then Riley stumbled into two road blocks. One was simple physics. As famously summed up by Calvin's dad in Calvin and Hobbes, records require a turntable's needle to travel at increasingly slower speeds as you get closer to the middle of the disc. Riley couldn't figure out how to get his prototype RokBlok to slow down as it went along. Then there was another problem: His idea wasn't as unique as he thought. "A friend of mine told me about the Soundwagon, this little toy VW bus made by a Japanese company called Tamco in the '70s, that could play records by driving on top of them," he said.
Riley despaired that his idea wasn't original, until he realized that the Soundwagon had a big design problem. When speakers spin, as they do in the Soundwagon, they create the doppler effect, best experienced in the caterwauling of an ambulance siren. But by using wireless to pipe the RokBlok's sound to a Bluetooth speaker, the RokBlok could bypass that particular issue. Not only that, but Riley realized that he could use the Soundwagon to backwards-engineer a solution to the RokBlok's slow-down problem. It turned out that Tamco's engineers had modified the gadget's resistor to essentially provide less electricity to the Soundwagon's wheels over time, allowing it to sync up to the speed of the record. He's now using basically the same solution.
Now on Kickstarter starting at $59, Riley admits that the RokBlok's not going to replace the turntables of dedicated audiophiles. But especially when paired with a decent Bluetooth speaker, it's still a nice way to spin records no matter where you are, with no risk to the records themselves... Though you might want to spare your most vintage, hard-to-find vinyl, just in case. You can preorder it here.
Steve Jobs hated it, but Ive's Twentieth Anniversary Mac set the stage for everything to come.
Perhaps you heard. A few weeks ago, Apple released a book called Designed by Apple in California: a $300 hardcover containing 450 photos of Apple products, chronicling almost 20 years of Apple's industrial design—and specifically, Jony Ive's Apple industrial design.
Apple sent me a review copy. My first impression? At almost 12 pounds, Designed by Apple in California is Apple's heaviest product this side of a 27-inch iMac. Seriously, there are weaponized toilet tank lids that weigh less than this thing.
But I also thought where the book started was interesting: 1998's candy-colored iMac G3, the first Mac released after Steve Jobs came to Apple. This is an important computer, true. But it's not the first computer Jony Ive designed for Apple. Nor is it his most forward-looking.
That honor would go to the Twentieth Anniversary Macintosh, sometimes shortened as the TAM, first released in 1997 at the eye-watering price of $7,499, or about double what a Mac Pro will cost you today. Only 12,000 were ever made, making it a collector's item to this day.
What's amazing about the TAM is that it foreshadows a lot of the work Ive would do on the Mac line over the next two decades. Like the current iMac, it's a slim, flat-screen all-in-one, released during an era when computers were distinctly cube-shaped. Everything's integrated: the screen, the speakers, the disc drive, and so on. It had an LCD screen, five years before they came to the rest of the Mac line, and weighed half as much as the iMac G3, released the following year.
If Apple under Ive has been defined by its obsessive pursuit of ever thinner, lighter devices, then the TAM was ahead of its time. But it's interesting to see other future Apple innovations the TAM took a stab at. Consider, for example, the fact that the TAM shipped with a removable trackpad that could be used alongside the keyboard, 13 years before Apple unveiled the Magic Trackpad.
It's true, the TAM doesn't share the same material sensibility as later Ive designs. It's mostly plastic, but that was typical of the era: Apple wouldn't really switch over to aluminum and glass as its primary materials for another decade. Yet, in this promotional video, a young, skinny, and curiously unshaven Ive delivers a very familiar sounding ode to the care his team put into the TAM's material process: The die-cast metal foot, for example, or the addition of metallic flaking to the plastic lacquer used in the case, all to give the TAM a premium feel and luster.
In any design retrospective of Ive's time at Apple, the TAM is an important computer to mention. So why is it missing from Designed by Apple in California?
Speculatively, there are two possible reasons why the TAM didn't make the cut (Co.Design reached out to Apple to find out more, and will update this post when we receive a response). First of all, Ive designed it before Steve Jobs took back the company; everything else in the book is decidedly post-Jobs. Second, Jobs himself reportedly hated the TAM, and made one of his first acts upon coming back as CEO the product's discontinuation. (This was also the period during which he briefly considered firing Ive.)
Outside of that, though, I think a big reason why Ive might have decided not to include it in Designed by Apple in California is simply that it doesn't share the same design language as the rest of the products in the book. It feels like a prototype, while the other products feel fully formed; a compromised iteration of Ive's grand vision pushed through a corporate culture that wasn't yet ready to truly let Ive put industrial design first and foremost.
That's an understandable perspective, but it's still a shame that it's missing from Designed by Apple in California. Because as Ive points out with all of his typical, soft-spoken, navel-gazing eloquence in the TAM promotional video above: "While [the TAM] doesn't have the answers for all the questions it asks, it's an important product . . . " It's the thesis of so much of Ive's work at Apple to follow.
Because if you can't give the gift of travel this Christmas, at least you can give the gift of traveling well.
Americans make over 405 million business trips per year—and that number is steadily increasing. But traveling, especially by air, hasn't gotten much easier, between cramped seats and endless security lines. Here are five gifts that cut down on some of the more annoying aspects of frequent travel—and one wildly expensive folly for non-business adventures.
GrandRevølution Washingtons
An Oxford that fits like a Nike sneaker? That's what Cole Haan have achieved with the GrandRevølution Washingtons, transforming the notoriously inflexible wingtip into a comfortable performance shoe, all without compromising its handsome looks. In fact, if anything, Cole Haan's GrandRevølution Washingtons look better than your usual pair of Oxfords. $400 will buy you a pair.
Louis Vuitton New Luggage
When you're Marc Newson, designer of special projects at Apple and legendary industrial designer in his own right, you know what you want in a stylish roller suitcase. So he teamed up with luggage legend Louis Vuitton to create his own stylish, colorful carry-on. It follows an Apple-like formula: the New Luggage Collection is 50% thinner and substantially lighter than Louis Vuitton's previous rollerbags, but can pack 13% more luggage. Want one? Expect to pay: prices range from $2,660 for the carry-on to $5,900 for the most luxurious stowaway. Buy one here.
Smart Coat
How thick does a coat need to be to keep you toasty in winter? If you buy the Smart Coat, not very. Emel+Aris's beautifully cut cashmere coats combine a slim outer jacket with a detachable vest that heats. At a price starting at around $1,200, they aren't cheap, but buy a Smart Coat, and you'll do more than save yourself from bringing more than one coat on your next business trip. You'll solve your heat regulation issues for good.
Away Bag
The biggest problem at an airport is wheeling your carry-on from one gate to the next. The second biggest problem? Keeping your gadgets charged while you do so. The brainchild of two Warby Parker alums, the Away bag minimizes the pain points both ways. A charger and USB battery integrated into the suitcase make it easy to keep juiced, while a number of considerate features like a built-in compression divider and laundry bag make this a truly convenient bag for the road warrior. Buy one starting at $225 here.
Water-Repellant Umbrella
Carrying a dry umbrella isn't much of a pain point, but a wet one? That's where things get annoying. So while Osaka-based designer Kazuyo Koike's umbrella might not be the smallest or most collapsible one on the market, it's probably the one you want, thanks to a moisture-resistant outer shell that makes it simplicity itself to shake off rain. A design so good, the MoMA sells it, you can buy one for $48.
Airstream Basecamp
Got someone in your life who wants to hit the open road? The Airstream Basecamp is for today's Sal Paradise: a teardrop of silver you can tow behind a small SUV with the perfect combination of Nuclear Age style and 21st-century technology. The perfect update to the classic trailer, the Airstream Basecamp is expensive at $35,000, but hey—imagine how great it'll look under the Christmas tree. Find one to buy here.
Amazon created its empire by making online shopping easy. Now, it's trying to do the same thing in the real world. It has just opened a grocery store in Seattle with no checkout line and no cashiers. Instead, you walk in, grab what you want, and walk out. Amazon automatically knows who you are and what you took, so once you leave, your Amazon account is automatically billed.
So how does it work? Hey, no one knows, and Amazon won't really say. Also, right now, only Amazon employees can shop there: even the press isn't being let in. (Amazon says Go will be open to everyone starting early next year). All we know is that it requires a special Amazon Go smartphone app, and that the House of Bezos is using some combination of machine learning, computer vision, and sensors to keep track of the items that customers pick up.
Speculatively, though, it appears to work like this. When you scan your Amazon Go app at the turnstiles, it logs you in via NFC, the technology Apple Pay uses to let you pay with your smartphone. This initial scan essentially gives Amazon permission to track you and bill you for what you take from the shelves. From there, Amazon starts tracking your face with ubiquitous video cams throughout the store, essentially tying your physical identity to your Amazon account. From there, it uses computer vision to detect which sections of the store you visit and where you take your products from. When you walk through the turnstiles again, it recognizes your face, logs you out, and bills your Amazon credit card.
Even if this isn't exactly the way the technology works, the effect is the same: it means that when you shop at an Amazon Go store, you'll be under constant video surveillance, with AIs poring over your every move to analyze how you shop. Considering the fact that Instagram and other networks will already target you with ads for Amazon items you literally looked at once, the world's largest e-retailer will soon be able to gather just as much data on how you shop in meatspace as it does online.
Is the convenience of walking into a store and not having to deal with a clerk worth giving Amazon that level of scrutiny into your meatspace consumer habits? Possibly not, but when Amazon in its video says it has spent the past four years asking what shopping would look like if you could just walk in, grab what you want, and go, that's what the company really means: it has spent four years figuring out how to start tracking you in the real world as effectively as it does online.
If that makes you uncomfortable? Keep waiting in line.
The machine is designer Sooji Lee's attempt to bring craftsmanship back to the age of digital reproduction.
Press down a letter on your keyboard, and your computer will fill your screen with it, in any font you want. It's a process we take for granted, says Design Academy Eindhoven graduate Sooji Lee, who wondered what it would take for a human to reproduce type—or at least one typeface, Bodoni—as reliably as a computer does.
For her project "How To Write Bodoni Lower Case," Lee created an enormous, pedal-operated wooden machine, which resembles an industrial loom in some respects. Totally analog, the machine simply exists to allow Lee to recreate Bodoni consistently. By changing the settings on her machine and pushing the pedal down, she can draw any lower case letter in Bodoni. It takes Lee some 74 steps, 43 presses of the pedal and 30 minutes just to reproduce the letter a. Other letters are simpler. Even so, it requires an incredible amount of patience to write anything on Lee's Bodoni machine.
Which is, of course, the point. Lee created her machine as part of a master's thesis on craftsmanship, to see what it would take to adapt the analog techniques of old world masters to the digital age. Craftsmanship requires patience and care, qualities in short supply when we're picking a font for our Word documents. Yet the fonts we're picking so carelessly are often extraordinary works of craftsmanship in their own right; "How To Write Bodoni Lower Case" posits that maybe if we're going to use them, we should do so with a modicum of the care that went into creating them in the first place.
The Infinite Galaxy Map can be added to forever, or at least until you run out of table space.
The universe is infinite, at least on a human scale. Comparatively, most jigsaw puzzles only feel infinite, especially when you're forced to do one at a family gathering. But the Infinite Galaxy Puzzle actually is endless. The puzzle, which was designed by the MIT alums at generative design studio Nervous System (no stranger to jigsaw puzzle design), tiles continuously so you can literally keep adding to it forever.
The Infinite Galaxy Puzzle features a NASA satellite photograph of the center of the Milky Way, but that's not what makes it special: it's the shape of the pieces. They're modeled after a dissected Klein bottle, a mathematical concept first described in the 19th century. This concept describes an infinite surface which, when traveled across in a straight line, eventually puts you back where you started, except flipped upside down.
What this means is that each one of the 133 pieces in the Infinite Galaxy Puzzle can be placed on either side, if you only flip it upside down. Consequently, the Infinite Galaxy Puzzle has no edges as we usually understand them in a jigsaw puzzle: any piece can be an edge piece, so you can never put it together the same way twice. It also means that if you wanted to, you could buy multiple Infinite Galaxy Pizzles, and keep on tiling them forever.
Originally available in a small production run, the Infinite Galaxy Puzzle quickly sold out. The next batch is now available for preorder from Nervous System for $100 apiece, and will begin shipping in early 2017.
But AIs aren't as smart as people. Not yet. That's because AIs lack general intelligence. They can't apply what they've learned from one problem to another—which is why even the best AIs are idiot savants: really smart in one arena, and dumb as sticks in all others.
So how can AIs reach this elusive general intelligence? OpenAI—an artificial intelligence research nonprofit backed by Microsoft, Elon Musk, and Peter Thiel—thinks it involves AIs playing video games. Thousands of video games. And that's just the beginning.
OpenAI's new software platform, Universe, allows AI agents—any artificial intelligence that is programmed to act upon its environment and work toward goals—to test itself on free Flash games, many provided by the Flash game portal Newgrounds, as well as more popular games like Grand Theft Auto V, Portal, World of Goo, RimWorld, Mirror's Edge, and more. Basically, it works by allowing AIs to connect to a virtual machine installation of a game and control it like a person, scanning the pixels to determine when it is winning and when it is losing. Better yet, it does it without any special programming on the developer's part for an agent to play a game: All an AI needs to be able to do is use its "eyes," just like a person. An agent can even keep track of its high score, using that as a reward to gauge its own performance. It's like an obstacle course for training AI.
Since the Universe platform essentially just gives AI agents a framework to control virtual machines—which can run any operating system or software—the future of Universe isn't limited to games. It can technically run any software, including productivity apps and websites. So programmers won't necessarily just train their agents on Grand Theft Auto V; they'll train them on Excel, Amazon, and Photoshop. Android support also works, and iOS is coming soon. Anything that runs on a computer is something that Universe can potentially train an AI on, and anyone—or at least, anyone with an AI—can access it for free.
"It's not just games," says OpenAI's Catherine Olsson, an alum of MIT's Brain & Cognitive Science group. "Our goal is that anything a human can do on a computer, an AI agent should also be able to do."
So is the point to create a single super AI, one robot to rule them all? Not really, says Jack Clark, communications director at OpenAI. Rather, it's to give multiple companies and institutions working on machine learning the ability to test whether or not their agents have general intelligence. "Look at this way," says Clark. "If I asked you: How would you characterize your intelligence? No matter your answer, it would be totally subjective." In a way, Universe removes the subjective guesswork, because if someone develops an AI that can play video games as well as it can book a flight, "it must have general intelligence." And if it has general intelligence, an AI is on the path to being as good as a person.
That opens the door to all sorts of possibilities. Digital assistants who are as good as people at scheduling your tasks, even if you've never completed a particular task before. Software that can test websites, or provide UX insights for designers and developers. Self-driving cars that can navigate traffic patterns they have never encountered. Even AIs that can adapt alongside humans, instead of just falling apart the second you ask them for something they don't know how to do. In other words, it's a step toward flexible computer intelligence.
Universe won't teach AIs to do these things by itself. Computer researchers will still need to figure out how to program them to be as good at learning as humans. But it will provide the obstacle course. And if everything goes well? It could help establish the first truly "human" AIs, although it's worth noting that there are many—including one of OpenAI's backers, Elon Musk—who worry that AIs that are as smart as humans might prove uncontrollable, and therefore, dangerous. HELLO PROFESSOR FALKEN. SHALL WE PLAY A GAME?
