Using building materials to monitor for high enriched uranium

A new paper details how small samples of ubiquitous building materials, such as tile or brick, can be used to test whether a facility has ever stored high enriched uranium (HEU), which can be used to create nuclear weapons. The technique could serve as a valuable forensic tool for national or international efforts related to nuclear nonproliferation and security.

“We can now use the housing structure itself as part of any nonproliferation monitoring efforts,” says Robert Hayes, an associate professor of nuclear engineering at North Carolina State University and author of the paper. “This work details the theory to test building material samples to differentiate between the forms of uranium used in nuclear power and the HEU that’s used to develop nuclear weapons.”

The technique builds on previous work done by Hayes and his research team.

The method requires testing a relatively small core sample of the relevant building material, about the size of your pinkie finger. The testing is done using hardware somewhat similar to that used to assess radiation exposure of dosimeter badges worn by workers in the nuclear power industry. In a sense, a small piece of any wall effectively becomes a dosimeter badge.

“Our technique allows us to determine how much radiation a material has been exposed to, in addition to the very types of radiation a material has been exposed to,” Hayes says. “Because different radionuclides have different radiation fields, these measurements allow us to determine which nuclear materials were stored near whatever building material we’re sampling.”

While this technique is new, there is already interest in it among the agencies responsible for nuclear monitoring — and Hayes is working to improve the technique further.

“We’re optimistic that this will be a valuable tool in the nonproliferation monitoring toolbox, but we need to address some existing questions,” Hayes says.

“For example, the radiation signature will vary depending on where the nuclear material was stored in relation to whatever sample we’re testing. If our sample was from brick that was right under a uranium storage container, the signature will be different than if the container was located 20 feet away, horizontally. Theoretically, these properties of the signature would be consistent over any gridded array of the same building material. Sampling such an array would then allow us to reconstruct not only what material was stored at a site, but precisely where it was stored. That’s something we’re working on now.”

The research was done with support from the Nuclear Regulatory Commission, under grant NRC-HQ-84-14-G-0059; and from the Consortium for Nonproliferation Enabling Capabilities under grant DE-NA0002576, which is based at NC State and sponsored by the National Nuclear Security Administration.

Story Source:

Materials provided by North Carolina State University. Original written by Matt Shipman. Note: Content may be edited for style and length.

Read More

Buildings made out of aerogel could help us survive on Mars “in our lifetimes”

Buildings made out of aerogel could help us survive on Mars “in our lifetimes”

Mars is a cold, inhospitable planet. Temperatures routinely drop to as low as -81 °F and down to -200 °F in some spots during winter. Meanwhile, the ultraviolet radiation could make humans seriously ill. So while science-fiction writers have imagined the possibility of terraforming the planet—of somehow reengineering the entire atmosphere so that it can host plants, farms, and people like Earth—scientists know that if we want to survive on Mars in the near-term, we’ll need to live and nurture life indoors.

New research published in Nature Astronomy explains how, by proving that growing food on Mars might be possible by building greenhouses out of a special material we already have on Earth: aerogel. “We’re excited about it, because we could potentially make portions of Mars habitable in our lifetimes,” says Robin Wordsworth, the Harvard researcher who led the study.

Thanks to Wordsworth’s science, we’ve learned it might be possible to build habitable structures on Mars. But intriguingly,  the research could help us build in some of the most extreme environments of our planet, too.

Aerogel on the hand of a researcher. [Photo: Science Photo Library/Getty Images Plus]

What is aerogel?

Aerogel was first discovered in the late 1980s. It is a clear material that, just like glass, is usually made from silica, explains Wordsworth. But aerogel’s molecular structure is far looser and more dispersed than glass. In fact, aerogel is up to 99.8% air, making it one of the least dense materials we know. That also means it’s remarkably light. If you had 150 bricks made of aerogel, they’d weigh the equivalent of a single gallon of water. Think of it like a translucent styrofoam.

Aerogel’s molecular structure also makes it a natural insulator. “If you compare to ordinary glass . . . silica aerogel is 10x more effective at insulating,” says Wordsworth. It also blocks nearly all infrared radiation coming its way, while allowing near all visible light through. As Wordsworth explains, “It really is a unique material,” as no other material juggles light in this same way.

Add all of this up, and you have a perfect material for cold environments—all the benefits of a greenhouse with all of the benefits of advanced heat insulation. Plus it’s light (so it would be feasible to bring it from Earth), and it protects from UV rays, which can damage the DNA of plants as well as humans. Given all these properties, it’s feasible, at least on paper, that we could build aerogel greenhouses on Mars.

What could aerogel do on Mars?

In Wordsworth’s test, a small piece of aerogel was set up in a lab under the same light and temperature conditions we’ve found on the red planet. “The idea was to . . . do a proof of concept that it benchmarks as well as expected in theory,” he says. What they found was that the aerogel performed admirably: A mere 2.5-centimeter-thick sheet of aerogel could raise the temperature under it by up to 150 degrees Fahrenheit. That’s warm enough to melt ice on Mars—and grow plants.

So, what would these Martian buildings look like? Wordsworth notes that aerogel itself looks something like solid smoke. Looking at an aerogel ceiling would be like staring at the sky on a cloudy day. But he doesn’t get into a lot of additional detail into how aerogel architecture would work. At minimum, aerogel could be laid down in sheets right on the surface of Mars, to melt existing ice and allow algae or aquatic plants to grow underneath.

Aerogel still has some mechanical issues that limit its use. While NASA actually used some aerogel to insulate the inside of the Mars rover and other high budget insulation projects, it simply hasn’t been proven out as a significant or ubiquitous building material yet. “There are some challenges that need to be worked on further,” says Wordsworth. “It’s a fairly fragile material. For its density, it’s amazingly strong, but it fractures easily. It’s not very flexible. You might want to combine it with another material in layers.” It’s possible that aerogel could be incorporated into more forgiving plastics, too, he says.

The most effective aerogel is made in tiles for various industries today, but Wordsworth’s team is beginning to consider how smaller, disc-shaped aerogel could perform as an insulator. A disc could create a more multifaceted surface, so the aerogel could be more adaptable to various architectural shapes.

The insulating qualities of ultralow-density aerogel are demonstrated under intense heat from a blowtorch.[Photo: © Roger Ressmeyer/Corbis/VCG]

Could we use aerogel to build on Earth?

If aerogel is so wondrous, why then don’t we use it to build greenhouses here on Earth? “Good question—it’s more expensive than everyday glass, but I think it’s also true you could burn your plants,” says Wordsworth. “There might be interesting applications thinking about habitats in extreme environments like Antarctica.”

Specifically, Wordsworth would like to do “small scale experiments” with aerogel in Chile’s Atacama Desert or Antarctica’s McMurdo Dry Valleys, a pair of very cold, very dry environments on our planet that are the most Mars-like areas of our planet. “Aerogel could have potential for larger buildings situated in those extreme environments [on Earth],” he says.

In other words, aerogel isn’t an architectural wonder material yet. But as scientists like Wordsworth conduct more tests and prove out new ways to use it, don’t be surprised to see it find its niche in the most extreme environments—from Earth to Mars.

Read More

An optimistic view of deepfakes

An optimistic view of deepfakes

Sunny Dhillon
Contributor

More posts by this contributor

Deepfakes are having a moment.

Their dangers are becoming more known and understood. The media is rife with articles detailing the speed at which the technology has grown in sophistication and become more accessible, as well as the risks involved.

Good.

The negative implications of deepfakes are troubling, and the better we understand them, the better we’ll be able to prevent their worst consequences. For better or worse, the technology is here to stay. But there is a “better” here—deepfakes have much in the way of lighthearted upside. 

Though the debate around deepfakes has grown in stature and complexity, we still struggle to agree on a definition of deepfakes. I think of it as any mimicry, manipulation, or synthesis of video or audio that is enabled by machine learning. Face-swapping, body puppetry, copying someone’s voice, and creating entirely new voices or images all fall into this category. Your Photoshop efforts, valiant though they are, don’t.  

Image synthesis and manipulation can be a powerful tool for creators

Visual storytelling is an expensive business. Hollywood studios spend billions on creating spectacle that wows their audience or transports them to another world. The tools they use to do so—the tools these big players use to close the gap between what they can imagine and what they can create—remain prohibitively expensive for most creators, though less so than a decade ago. Deepfake tech incorporates the ability to synthesize imagery, potentially giving smaller-scale creators a similar capacity for bringing imaginative creativity to life.

Synthesia is a company with a commercial product that uses deepfake tech to do automated and convincing dubbing through automated facial re-animation. They shot to prominence with a video that featured David Beckham talking about Malaria in nine languages, but their product could also be used to expand the reach of creators around the world. If you’re a talented artist who isn’t working in one of the world’s dominant languages, it’s potentially career-changing to have access to a product like this, which could make your work viable in additional languages and countries.

Adobe VoCo is software — albeit still at a research and prototyping stage — that makes it easier for creators to produce speech from text and edit it the way they would edit images in Photoshop. So if you want your movie short to be narrated by Morgan Freeman, you might be able to make that happen.

Tinghui Zhou, the founder and CEO of Humen, a company that creates deepfakes for dancing, sums up the industry’s goals: “The future we are imagining is one where everyone can create Hollywood-level content.” (Disclosure: I am an investor in Humen).  

In the same way YouTube and Instagram shrunk the distribution and creation advantage that entertainment companies and famous photographers enjoyed over talented amateurs and enthusiasts, this bundle of technologies might diminish the production advantage currently possessed by big budgets and visual effects houses.

Mimicry and manipulation of real life have always been part of art.

The applications mentioned above are all to do with closing the gap between creators with different resources, but deepfake tech could also enable entirely new forms of content that rest on the ability to mimic and manipulate material. Every medium of entertainment has incorporated the stretching, reflection, contortion, and appropriation of real source material for the purposes of entertainment. 

We can already see the evidence of these new applications in the still-nascent use of deepfake tech today. While face swapping for porn lies at the malicious end of the spectrum, more benignly the technology’s introduction also sparked a wave of face swapping Nicolas Cage into different movies.

It might seem banal, but it was a form of content creation that, while previously technically possible, was practically infeasible before deepfakes. It’s not hard to imagine that the next deepfakes content craze will be driven by automated lip-syncing, dance mimicry, or celebrity voice impressions.

Respeecher and Replica.AI are just two companies making voice mimicry accessible to non-techies. Check out my demo with Replica’s tech in San Francisco a few weeks ago (recognize the voice?). It’s a small slice of the future of entertainment and content. If you believe that culture in the digital era is the culture of remixing, then deepfake tech has an important part to play in the creation of that culture. 

Deepfakes bring us closer to believable virtual humans

The ability to mimic faces, voices, and emotional expressions is one of the most important steps toward building a believable virtual human that we can actually interact with. We’re already taking tentative steps down the path to virtual humans. Personal assistants like Alexa, Siri, and Cortana have been around for several years, reached a tipping point of consumer use, and are quickly improving. Having said that, in 2019 they still feel more like a new user interface you have to pass precise instructions to rather than a virtual being you can interact with. Think a command line operated by speech. 

Virtual humans are entering the mainstream in a different way: Through the recent wave of digital influencers. I previously wrote about this trend in the context of animation history, but digital influencers are also meaningful in the context of believable virtual humans. Digital influencers operate on the same planes of interaction — think your Instagrams and Pinterests — that most people do.

As such, you and I can comment on a Lil Miquela post or message Astro. This is interaction with a being that isn’t real. The digital influencer isn’t really responding to you in their own words — their content is created by storytellers, much as Pixar films have writers. But these digital influencers are laying the social groundwork for interaction with true virtual beings.  

Deepfakes have the potential to plug the technological holes in smart assistants and digital influencers. Pushing Alexa or Lil Miquela to the level of virtual humans like Samantha from Her or Joi from Bladerunner 2049 requires the capacity to encompass and express human body language, speech, and emotion. If we counted the number of unique combinations of pose, vocal nuance, and facial expressions you’ve made in your lifetime, it would likely number in the billions. For virtual humans to be believable, their actions can’t be preprogrammed in a traditional hard-coded sense, but must instead be extremely flexible.

Deepfake tech typically takes tons of examples of human behavior as inputs and then produces outputs that approximate or elaborate on that behavior. It could grant smart assistants the capacity to understand and originate conversation with much more sophistication. Similarly, digital influencers could develop the ability to visually react in a believable way in real time, thanks to deepfake tech. Bringing Mickey Mouse to life beyond a Disney cartoon or guy in a suit at Disneyland is where we’re headed. 3D hologram projections of animated characters (and real people) that are able to speak in a realistic sounding voice, moving like their real world counterpart would. 

Creativity starts with copying. Elaboration follows duplication. It is no different with deepfakes, which will democratize access to creativity tools in entertainment, enable entirely new forms of content, and bring us closer to believable digital humans. That is why I think there is as much reason to be excited about the technology’s virtues as there is to be concerned about its vices. 

Read More

Pay what you want for an awesome eBook bundle of Arduino projects!

Pay what you want for an awesome eBook bundle of Arduino projects!

The easiest way to get into hobby robotics is to learn how to create and code your projects using Arduino. You can learn and follow along with starter projects right at home even if you’re new to coding or robotics. You just need the right reading material to teach and inspire you with fun Internet of Things projects

Teach yourself basic robotics with these arduino eBooks!

Right now, Android Central Digital Offers has a pay what you want deal available for the The Complete Arduino eBook Bundle. Instead of paying $41 for each eBook individually, you can get all six available ebooks just by paying more than the average price — currently around $15 right now! If you’ve always been interested in circuity and creating your own electronic and robotic builds, now is the time to take the plunge.

If you pay less than the average — even just $1 — you’ll get Python Programming for Arduino, a 400-page book that introduces you to how Python and arduinos can be used to create all sorts of great IoT projects. The rest of the books are unlocked by beating the average price, and for around $15 that’s a great deal:

  • Mastering Arduino
  • Building Smart Drones with ESP98266 & Arduino
  • Arduino for Kids
  • Internet of Things with Arduino Cookbook
  • Arduino Wearable Projects

You might be inspired Just from reading those those titles. If you were, then you need to get this bundle today!

Become an Arduino pro now!

Read More

Founder of Neo-Nazi Site Daily Stormer Ordered to Pay $14 Million to Target of Racist ‘Troll Storm’

Founder of Neo-Nazi Site Daily Stormer Ordered to Pay $14 Million to Target of Racist ‘Troll Storm’

Tanya Gersh hugs her father after the hearing on July 11.

Tanya Gersh hugs her father after the hearing on July 11.
Photo: AP

Tanya Gersh has faced anti-Semitic harassment since late 2016—calls, tweets, emails, and texts demeaning her and saying she was going to die—upending the once-peaceful life she had in a quaint Montana town.

The barrage began soon after Daily Stormer founder Andrew Anglin shared Gersh’s contact information on his neo-Nazi website and told his followers to go after her.“Let’s hit ‘em up,” he posted. “Are y’all ready for an old-fashioned Troll Storm?”

Anglin’s call to arms came after Gersh was involved in a neighborly dispute with the mother of white supremacist Richard Spencer following news that a public protest was planned at a building owned by Sherry Spencer.

According to BuzzFeed News, Anglin also posted an article on his site titled “Jews Targeting Richard Spencer’s Mother for Harassment and Extortion—TAKE ACTION!” in which the Anglin wrote, “If you’re in the area, maybe you should stop by and tell her in person what you think of her actions,” but added that the action shouldn’t be violent.

On Monday, federal judge Jeremiah C. Lynch ruled that Anglin should pay more than $4 million in compensatory damages and $10 million in punitive damages, which is the maximum under Montana state law, reports NBC News. Lynch stated that Anglin should get the maximum punishment due to “the particularly egregious and reprehensible nature of Anglin’s conduct.” Lynch’s ruling must still be approved by a U.S. district judge.

According to NBC News, Lynch also called for a permanent injunction that would force Anglin to take down all content related to Gersh, as “the atrocious conduct directed at Gersh and her family has not entirely abated.” As of Monday morning, the Daily Stormer still still hosts material about Gersh.

Anglin seems to be in hiding and did not appear for a deposition in April. At the time, Anglin’s former lawyer Marc Randazza told the Missoulian his relationship with Anglin had “broken down.”

“When a federal judge tells you to do something and you refuse, you put your lawyer in a difficult position.” Randazza told the Missoulian. “My client made the decision years ago he was going to expatriate himself and never return.”

Gersh was represented in part by the Southern Poverty Law Center, which filed the federal suit in 2017. According to SPLC, Gersh and her family received more than 700 messages harassing them before the group filed the complaint. In a release published following Lynch’s ruling, SPLC wrote that Anglin planned an armed march that was supposed to end at Gersh’s house. The march never took place but Anglin reportedly advertised the event with an image of Gersh and her son superimposed onto a photo of the Auschwitz concentration camp.

“This lawsuit has always been about stopping others from enduring the terror I continue to live through at the hands of a neo-Nazi and his followers, and I wanted to make sure that this never happens to anyone else,” Gersh said in a statement shared by SPLC. “A clear message has been sent to Anglin and other extremists: No one should be terrorized for simply being who they are, and no one should ever be afraid for being who they are.”

Read More

Google’s six rules for great data design

Google’s six rules for great data design

Google does everything from building smartphones to creating driverless car companies. So it’s easy to forget that its official mission is “to organize the world’s information and make it universally accessible and useful.” Google is and will always be a company anchored in the business of gathering data and serving it up in a comprehensible way, whether that’s through Search results, Google News alerts, or restaurants you might like automatically popping up on Google Maps.

So it’s perhaps no surprise that Google now wants to help others package their data, too. Manuel Lima—a celebrated information designer who founded and heads Google’s data visualization team (formed in 2018)— has published Google’s Six Principles for Designing Any Chart, a cheat sheet to clear, accessible data visualizations. The principles weigh in on topics ranging from typography to the practice of dishonestly presenting data to serve an agenda. Additionally, Google has launched a series of guidelines to design better graphics, as part of Material Design, the company’s design language for user interfaces.

[Image: courtesy Google]

Google has more deep data knowledge than any company in the world, and it is no slouch in the discipline of design. It’s only natural that the company would combine this expertise. Initially, the audience for the new data design guidelines was Google itself, but much as it did for Material Design, the company decided to publicize these best practices and encourage others to adopt them—anyone from app developers to everyday people who are left wondering why their PowerPoint chart sucks.

“We started doing this internally as a way to guide [employees] through the do’s and don’ts of chart creation,” Lima tells Fast Company. “After conducting various research studies and partnering with teams across the company, the do’s and don’ts evolved into a set of high-level principles that were strongly rooted in Google-wide tenets crucial to the company’s growth, brand, and culture. These principles are meant to be generative and not prescriptive. We hope they can help any chart creator during ideation and evaluation.”

[Image: courtesy Google]

The six principles read something like an introductory data design course. “Don’t distort or confuse the information for embellishment or partiality,” reads one tip. “Respect different user needs on data depth, complexity, and modality,” reads another. To approach data with journalistic rigor, and to make a graphic that is both glanceable for casual use and deep enough for experts, are the sorts of insights that are well known within the (relatively small) data visualization community, but potentially not outside of it.

Meanwhile, the full guidelines walk you through several specific examples of good data visualization, illustrated with charts built by the Google data viz team itself (in this way, it is similar to a 2018 project called From Data to Viz). The guidelines identify charts you’ve seen but probably never named, like “candlesticks” and “waterfalls,” while breaking down contexts where different charts makes sense.

[Image: courtesy Google]

Particularly useful are those aforementioned “dos and don’ts.” They are shown side by side to illustrate common mishaps in chart creation, like line charts rendered in so many colors they look the scribbles of a 5-year-old wielding a big box of Crayola, or graphs that love bold typography so much that EVERYTHING BECOMES EQUALLY, LOUDLY UNIMPORTANT.

[Image: courtesy Google]

One particular, if small, point that gives some insight into Google’s chart-design philosophy is its love for labels at the cost of cleaner graphics. “Avoid using solely icons and symbols to represent important information,” one tip reads. Does a graphic need to define the Wi-Fi symbol we all see every single day of our lives? To Google’s information design team, yes, yes it does. Clarity is prized above all else.

In any case, the guidelines are a worthwhile read to both professional designers and occasional PowerPoint chart users alike. You will undoubtedly learn something from the company that is organizing (and, ahem, collecting) the world’s information.

Read More

Pay what you want for an awesome eBook bundle of Arduino projects!

Pay what you want for an awesome eBook bundle of Arduino projects!

The easiest way to get into hobby robotics is to learn how to create and code your projects using Arduino. You can learn and follow along with starter projects right at home even if you’re new to coding or robotics. You just need the right reading material to teach and inspire you with fun Internet of Things projects

Teach yourself basic robotics with these arduino eBooks!

Right now, Android Central Digital Offers has a pay what you want deal available for the The Complete Arduino eBook Bundle. Instead of paying $41 for each eBook individually, you can get all six available ebooks just by paying more than the average price — currently around $15 right now! If you’ve always been interested in circuity and creating your own electronic and robotic builds, now is the time to take the plunge.

If you pay less than the average — even just $1 — you’ll get Python Programming for Arduino, a 400-page book that introduces you to how Python and arduinos can be used to create all sorts of great IoT projects. The rest of the books are unlocked by beating the average price, and for around $15 that’s a great deal:

  • Mastering Arduino
  • Building Smart Drones with ESP98266 & Arduino
  • Arduino for Kids
  • Internet of Things with Arduino Cookbook
  • Arduino Wearable Projects

You might be inspired Just from reading those those titles. If you were, then you need to get this bundle today!

Become an Arduino pro now!

Read More

Viking-style seaweed thatch updated into prefab panelling

Viking-style seaweed thatch updated into prefab panelling

Seaweed Thatch Reimagined by Kathryn Larsen

Traditional Danish seaweed thatching could be updated into a sustainable contemporary building material, according to research by Kathryn Larsen.

Copenhagen School of Business and Design student Larsen became interested in the potential for seaweed architecture after studying the Viking practice of thatching roofs using eelgrass on the island of Læsø.

Seaweed Thatch Reimagined by Kathryn Larsen

“Eelgrass is a fantastic material that is naturally fireproof, rot resistant, carbon negative and becomes entirely waterproof after about a year,” said Larsen. “It also insulates comparably with rockwool [a common type of insulation].”

“Plants grow in it, giving the effect of a green roof. It is a material that we should be considering seriously in an era of climate change.”

Seaweed Thatch Reimagined by Kathryn Larsen

A student of the Material Design Lab at the Copenhagen School of Business and Design (KEA), Larsen went on to test what a contemporary iteration of seaweed thatching might look like.

She tried different binders and grid spacing, eventually developing a set of prefabricated eelgrass-thatched panels to suit roofs or facades. Colloquially known as a seaweed but actually a type of seagrass, eelgrass is common across the UK and Scandinavia.

Larsen notes that while eelgrass levels are diminishing, there would be enough to supply a Danish construction industry, and that there are ways to farm and harvest the plant without harming the marine ecosystem.

She drew on the results of a 2013 project, The Modern Seaweed House, built on Læsø by architecture studio Vandkunsten and non-profit organisation Realdania Byg.

In contrast to the vernacular buildings on the island, where the seaweed is stacked into shaggy roofs about a metre thick, the Modern Seaweed House has a slimmed-down look, achieved by stuffing the eelgrass into net pillows around the facade.

She also built on the work of Copenhagen architects Studio Seagrass, who have made innovative use of eelgrass in interior design and other areas.

Seaweed Thatch Reimagined by Kathryn Larsen

Larsen’s eelgrass panels are now eight months into a one-year testing period on the KEA roof. At this point Larsen says they are “almost entirely intact” and beginning to grow moss.

The architectural technologist has secured funding to continue her research into seaweed thatching and build more prototypes in the year to come.

“I hope to test the panels u-values, to see what insulation properties they can bring to construction,” said Larsen of her future plans.

Read More

Facebook researchers propose an AI assistant for Minecraft

Facebook researchers propose an AI assistant for Minecraft

If you’ve ever wished Minecraft had an Alexa-like assistant that could perform any task asked of it, you’re in luck. Facebook researchers recently argued for an interactive, collaborative Minecraft bot for natural language understanding (NLU) research. They posit that the constraints of Minecraft make it well-suited to experiments in various NLU subfields, and to this end, they’ve made baseline data, code, labeling tools, and infrastructure freely available on GitHub.

Their work to an extent builds on LIGHT, an open source research environment in the form of a large-scale, crowdsourced text adventure within which AI systems and humans interact as player characters. Scientists at Facebook AI Research, the Lorraine Research Laboratory in Computer Science and its Applications, and the University College London detailed LIGHT in paper published earlier this year.

“Despite the numerous important research directions related to virtual assistants, they themselves are not ideal platforms for the research community. They have a broad scope and need a large amount of world knowledge, and they have complex codebases maintained by hundreds, if not thousands of engineers,” wrote the coauthors in a preprint paper published on Arxiv.org. “Furthermore, their proprietary nature and their commercial importance makes experimentation with them difficult. Instead of a ‘real world’ assistant, we propose working in the sandbox construction game of Minecraft.”

For those unfamiliar, Minecraft is a voxel-based building and crafting game with procedurally created worlds containing block-based trees, mountains, fields, animals, non-player characters (NPCs), and so on. Blocks are placed on a 3D voxel grid, and each voxel in the grid contains one material. Players can move, place, or remove blocks of different types, and attack or fend off attacks from NPCs or other players.

The researchers, then, describe a Minecraft bot that understands natural language commands (e.g., “build a tower 15 blocks tall and then put a giant smiley on top”) fed to it via the in-game chat window. They concede that implementing this is easier said than done, namely because of the complexity of tasks players might ask the bot to perform. In the aforementioned example — “build a tower 15 blocks tall and then put a giant smiley on top” — the assistant would need to understand the meaning of “tower” and “smiley” and how to build them; know that “15 blocks high” measures the height of the tower; recognize the significance of “15”; and reconcile the relative position “top”.

Still, the paper’s coauthors assert that Minecraft’s task space and environment have “regularities” that could be used to simplify task execution. For instance, sets of language/action templates for generating example task commands could be used to build training data and inform the structure of the bot’s underlying NLU models. Moreover, Minecraft’s structure could function as a knowledge resource shared between AI and player. For example, if a user asks the assistant to “build a smiley,” the agent could infer that “a smiley” is a kind of block object because “build” is a common task the bot would already understand.

The researchers make a case for a modular approach to streamline a hypothetical assistant’s design and subsequent research. They propose that the actions necessary to complete basic Minecraft tasks (like path-planning and building) could be scripted by accessing the game’s internal world state. Furthermore, they note that it’d be relatively easy to collect or generate data for actions by recording players’ interactions with the assistant.

Formidable challenges stand in the way of a Minecraft bot that’s “engaging” and “fun,” the team points out. It’d need to be immediately responsive to feedback, as latency often has a large effect on players’ impression of performance, and it’d have to “optimally” interact with players by seeking clarification without bombarding them with annoying questions. But despite the blockers, the team firmly believes that Minecraft is ideal for studying learning from interaction, and especially learning from language interaction.

“[I]nstead of [exploring] ML methods [that can] can learn representations of the environment that allow an agent to act effectively … we are interested in the problem of what approaches allow an agent to understand player intent and improve itself via interaction, given the most favorable representations … of the environment we can engineer,” wrote the team. “While we are sympathetic to arguments suggesting that we will be unable to effectively attack the NLU problems without fundamental advances in methods for representation learning, we think it is time to try anyway.”

Read More

Sidewalk Labs reveals the key to its smart city plans: An $80M building factory

Sidewalk Labs reveals the key to its smart city plans: An $80M building factory

Today, Google sister company Sidewalk Labs released a draft of its master plan to transform 12 acres on the Toronto waterfront into a smart city. The document details the neighborhood’s buildings, street design, transportation, and digital infrastructure—as well as how the company plans to construct it.

When a leaked copy of the plan popped up online earlier this year, we learned that Sidewalk Labs plans to build the entire development, called Quayside, out of mass timber. But today’s release of the official plan reveals the key to doing so: Sidewalk proposes investing $80 million to build a timber factory and supply chain that would support its fully timber neighborhood. The company says the factory, which would be focused on manufacturing prefabricated building pieces that could then be assembled into fully modular buildings on site, could reduce building time by 35% compared to more traditional building methods.

“We would fund the creation of [a factory] somewhere in the greater Toronto area that we think could play a role in catalyzing a new industry around mass timber,” says Sidewalk Labs CEO and chairman Dan Doctoroff.

[Image: Snøhetta/courtesy Sidewalk Labs]

However, the funding of the factory is dependent on Sidewalk Labs being able to expand its development plan to the entire riverfront district. The plan points out that Quayside’s 10 buildings are too few to make the factory’s $80 million investment—which will come from Sidewalk and other private partners who have yet to be determined—worth it.

Mass timber is an emerging building material that’s made out of pieces of wood laminated together to make them stronger. It has been hailed by many scientists and architects as the future of building because it is just as strong as steel, more fire-resistant than regular wood, and is environmentally friendly to boot: After all, a tree pulls carbon out of the atmosphere, while making steel and concrete generates emissions though some scientists debate that it’s better to leave trees in the ground). And the building material is growing in popularity: according to Sidewalk, there have been 21 timber towers taller than seven stories completed or under construction since 2013.

Click here for a larger version. [Image: courtesy Sidewalk Labs]

Sidewalk proposes sourcing spruce and fir trees from the forests in Ontario, Quebec, and British Columbia. While Canada has 40% of the world’s sustainable forests, Sidewalk claims, the country has few factories that can turn these trees into the building material. That’s why the company proposes starting a factory to process two kinds of mass timber: Cross-laminated timber (CLT) and glulam beams. The latter is meant specifically to bear the weight of the 30-story buildings Sidewalk hopes to build. While Sidewalk says that 84% of the larger district would be handed over for development by local companies, the plan requires that these companies uphold the same sustainability standards when it comes to performance.

Sidewalk says companies wouldn’t be required to build with CLT and glulam, but since the company’s reason for building the mass timber factory is that there aren’t many existing manufacturers to meet the needs for a full-scale development, the company’s plan might ultimately push any third-party developers toward using its factory to source materials.

Click here for a larger version. [Image: courtesy Sidewalk Labs]

The primary way that Sidewalk aims to reduce construction time is by designing the buildings to be completely modular, using a smaller kit of parts: “To accelerate project timelines, improve predictability, and reduce costs in a holistic way, Quayside’s buildings would draw from a complete library of factory-made building parts that can be customized for each project to allow for a diverse and interesting variety of buildings that achieve design excellence,” the plan states.

To demonstrate how this would work from a design perspective, Sidewalk Labs asked three architecture firms—Heatherwick Studio, Snohetta, and Michael Green Architecture—to use the same set of modular pieces to come up with building designs for different parts of the Quayside development. The plans, which were released in February, are all vastly different. Michael Green Architecture’s renderings reveal skyscrapers with wooden slats along their exteriors, while Snohetta’s feature a large curved facade paired with two towers. Heatherwick’s buildings have a layered, organic form. All of the renderings are just concepts for the time being, and the company says they do not reflect final design or engineering decisions.

Click here for a larger version. [Image: courtesy Sidewalk Labs]

The factory is also key to another of Sidewalk’s promises: Jobs. According to Sidewalk, the factory itself would create 2,500 jobs along the entire supply chain over a 20-year period.  But even if the Canadian government approves Sidewalk’s plan and commits to building out the entire waterfront district to take advantage of the mass timber factory’s economies of scale, there are other regulatory hurdles to overcome. Right now, the building code in Toronto doesn’t allow for timber buildings over six stories tall. All of Sidewalk’s proposed buildings are over six stories, and many of them go up to 30 stories. Doctoroff said he was optimistic that the company will be able to get regulations changed if the city decides to adopt the plan. There are several examples of timber buildings that are already under construction, with a planned skyscraper in Japan that will be 70 stories.

Sidewalk’s proposal is the result of 18 months of planning, which involved getting feedback from community members and prototyping elements like a building raincoat that the company hopes to include in the final development. It has come under fire from privacy advocates in particular, and the Canadian government is currently facing a lawsuit from a civil liberties group over its decision to allow a corporation to propose public privacy governance standards.

Now that the company has released the plan, it will be up to the Canadian government to decide whether to move forward. And the mass timber factory, in particular, will be dependent on the government adopting Sidewalk’s plan wholesale, far beyond the Quayside development—a reminder that Sidewalk is a corporation that’s here to make money, dangling investment dollars in front of the government to incentivize it to embrace Sidewalk as the developer for the entire area.

Read More