title
stringlengths 2
283
⌀ | author
stringlengths 4
41
⌀ | year
int64 2.01k
2.02k
| month
int64 1
12
| day
int64 1
31
| content
stringlengths 1
111k
⌀ |
|---|---|---|---|---|---|
Google responds to EU antitrust charges against Android
|
Natasha Lomas
| 2,016
| 11
| 10
|
Google has now filed its official response to the EU’s antitrust charges against its Android mobile OS. The company had been granted a deadline extension by the European Commission, which kicked off a formal investigation into Android back in . According to the EC is intending to hit Google with deterrent fines over its behavior with Android, and also in another antitrust case against Google Shopping. Among the complaints about Android that the EU Competition Commission has been investigating are — In a detailing the general thrust of its arguments, Google’s SVP and general counsel, Kent Walker, rebuts these charges, arguing the Android OS has created a thriving and competitive mobile app ecosystem. “The response we filed today shows how the Android ecosystem carefully balances the interests of users, developers, hardware makers, and mobile network operators. Android hasn’t hurt competition, it’s expanded it,” he writes, going on to argue that Google’s “voluntary compatibility agreements” for Android OEMs are a necessary mechanism for avoiding platform fragmentation which would make life harder for app developers. Walker claims the EC is overlooking a large chunk of relevant competition by not factoring in Apple’s iOS when it considers the mobile app ecosystem. However Android remains the dominant mobile OS in the region, with a marketshare of around 90 per cent as the EC sees it — hence the Commission’s focus on the platform. The EC is also factoring in Google’s regional dominance in general Internet search (circa 90 per cent of that market), which it has been investigating for some six years now — with the issue being how one dominant tech platform position might be being abused to cement another. “To ignore competition with Apple is to miss the defining feature of today’s competitive smartphone landscape,” argues Walker, also referencing iOS as a motivating factor in Google seeking to simplify developing for the Android ecosystem vs Apple’s closed source platform. “The Commission’s proposal risks making fragmentation worse, hurting the Android platform and mobile phone competition,” he adds. in the antitrust case argue that Google’s “ ” of Android makes it more akin to closed than open source. In a statement given to , FairSearch lawyer Thomas Vinje said: “The truth is that Android is today a closed operating system, and any claim to the contrary is disingenuous. Google imposes severe sanctions on those who defy its insistence on conformity.” Responding to criticisms of how it requires manufacturers to pre-load a suite of its own apps on devices if they want access to the Google Play store, Walker says Google does this to present a competitive offering vs rival platforms, with their own apps preloaded, while also arguing it’s easy for consumers to delete preloaded apps or download alternatives. Although the competition point is really about whether the discoverability and visibility of a preload on a dominant platform is an unfair advantage, not that other apps are being blocked at a technical level. He goes on to name-checking Spotify and Snapchat as two examples of third party app successes that have been possible despite Google’s own Android preloads. Lastly, Walker says Google requiring Android OEMs to use Google search by default is effectively its payment for offering device makers use of the entire suite for free, i.e. without a licensing fee. “This free distribution is an efficient solution for everyone — it lowers prices for phone makers and consumers, while still letting us sustain our substantial investment in Android and Play,” he argues. He also describes open source platforms as “fragile” and argues the Commission’s approach risks upsetting the “balance of needs” between users and developers, as well as suggesting their action could signal they favor “closed over open platforms”. But again here, complainants would argue that the way Google controls Android makes it akin to a closed platform at this stage. The EC can fine Google up to 10 per cent of its global revenue — some $7.5 billion — if it finds the company has breached European competition law. Google has already faced antitrust . Google also responded to the two other European Union antitrust charges — relating to its Google Shopping price comparison service, and its AdSense ad placement service. In the Shopping instance, Walker argued the online shopping ecosystem has evolved beyond price comparison sites — making them irrelevant in what he described as a vibrantly competitive market, focused on other channels for reaching consumers such as retailers’ own mobile apps. Regarding AdSense, Google said it has made some changes to how it operates the service — in the hopes of satisfying the Commission.
|
Google says there are now 2 billion active Chrome installs
|
Frederic Lardinois
| 2,016
| 11
| 10
|
Google is hosting its today. There hasn’t been a lot of news out of the event, but one number that stood out in today’s keynote by Chrome Engineering VP Darin Fisher was that there are now 2 billion Chrome installs in active use across desktop and mobile. This is the first time Google has shared this number. Sadly, Google didn’t announce any new user numbers for Chrome today. The latest stat for active Chrome users remains at 1 billion — a number Google shared in April. While this number is surely higher today than it was six months ago, the company decided to focus on the number of active browser install today. “I wanted to make this point that there are a lot of Chrome browsers out there,” Fisher said. “What’s exciting about this to you all is that when you think about building for the web, there’ a lot of browsers out there that implement the latest web standards — that implement the latest and greatest web features.” As Google likes to say, it now has seven products with more than a billion users (Gmail, Android, Chrome, Maps, Search, Youtube and the Google Play store).
|
Analysis of social media did a better job at predicting Trump’s win than the polls
|
Sarah Perez
| 2,016
| 11
| 10
|
Many of the polls forecasted an easy win for Democratic nominee Hillary Clinton in the U.S. presidential election, but we now know their prediction models were flawed. While data scientists are scrambling to figure out what went wrong, a number of social media analytics firms are touting the fact that their measurements did a better job of tracking the situation out in the real world, and saw the potential for Trump’s win well in advance. “Analysts monitoring the social media activity of both campaigns on the major social media channels saw the outcome of this election coming months ago, and kept talking about the massive silent voter base that was forming around the Republican nominee,” said Phil Ross, a social media analyst at . “Social media analysts continually sounded the alarm that all of the polls were not reflecting the actual situation on the ground in the pre-election landscape,” he added. Clinton outspent Trump on TV ads, set up more field offices, and sent staff to swing states earlier, but Trump simply better leveraged social media to both reach and grow his audience. It seems, too, that Trump benefitted from that old adage, “any press is good press.” While the October 7th leak of the Access Hollywood tape was reported as a low moment in the Trump campaign, it expanded his social media audience more than any other until Election Day, Socialbakers found. During the campaign, many analytics firms noticed that a growing number of swing voters were engaging with Trump’s messages on social media. Of course, at the time, it was unclear that social media engagement would translate into real-world votes that could put Trump in the Oval Office. After all, social media is often home to more passive participation when it comes to key matters – and routinely criticized for it. Pundits have as holding little meaning beyond raising awareness at best, and not often resulting concrete change. In a recent example, critics scoffed at the armchair activism of the Facebook Dakota pipeline check-in movement. A million after a viral post told them it would protect the protesters in North Dakota from police surveillance. (That necessarily the case). In other words, historically speaking, no one was sure that a like or re-share meant Trump was getting a vote. Especially as the polls told us otherwise. Outside of engagement alone, other firms like saw that social media sentiment towards Donald Trump trended more positively than sentiment towards Hillary Clinton on Election Day. Over the course of the campaign, Trump saw more mentions on social media associated with positive sentiment than his rival, as well. She closed that gap in late October and early November, but the good vibes for Trump soared once again just ahead of November 8th. Meanwhile, though Trump had trended more negatively than Clinton up until the final debate, negative sentiment was close for the two candidates as they headed into Election Day. A similar trend was spotted by the analytics firm , today , which found that Trump had more support on Facebook and Twitter from the beginning of October through November 7th. Positive sentiment for Trump was 58% to Clinton’s 48%, the firm said. And finally, also found that Trump had more Twitter mentions throughout most of the election cycle, including on Election Day. Between the start of November 8th and November 9th at 1:30am EST, Trump accumulated more than 4.9 million mentions while Clinton has just over 2.7 million. But not all social media analytics firms believed their data could point to a Trump win. A contrarian for example, noted how one firm, Spredfast, believed there were too many conflicting signs on Election Day for them to make a prediction. Still, even social media may have missed many Trump voters. During this election cycle, there were a number of silent Trump supporters, as well as those who were quietly leaning in his direction, who may have not made their views known on social media. It appears that at least some of these supporters only took to social media once a Trump win looked possible. After Trump took swing states Florida and Ohio, positive tweeting in support of Trump spiked, Spredfast told CNET. However, social media’s success likely means it will be given more attention in future elections. But it can’t see it all. That’s why we still need polls – preferably accurate ones next time.
|
Pinterest finds another way to track purchases via its new “Tried It” button
|
Sarah Perez
| 2,016
| 11
| 10
|
Pinterest has historically been more an idea board – a place to dream about ideas you want to try or products you want to buy. Today, however, the company is rolling out a new feature aimed at encouraging people to track and share the things they’ve actually done, too. With a new “tried it” checkmark on Pins, Pinterest users can share the ideas they’ve tried and would recommend to others. The service will also allow you to track the Pins others have tried in a new feed. The company says the feature will make it easier for people to find tips about things like recipes and vacation spots, or learn shortcuts for accomplishing projects, for example. It could also help to surface post-purchase product reviews – something that’s important to Pinterest as it has been attempting to move into a more actionable space over the years. For some time, the company – those that, when clicked, lead to product pages on e-commerce sites. But not all Pinterest users will end up buying products via the site. After all, Pinterest still serves more to inspire future purchases, rather than trigger instant gratification. Without being able to point to its ability to significantly increase conversions on shopping sites, to shift more of their budgets over to Pinterest. With “tried it” tied to products and other items across Pinterest, though, the company could have a better means of tracking Pins throughout the shopping cycle – from dreaming to doing, so to speak – outside of direct conversions from Buyable Pins. Using the new “tried it” checkmark is as simple as clicking the checkmark above the Pin. Pinterest also prompts you to offer feedback by asking “How’d it go?” You can then click on a happy-faced “Love It” button or a sad-faced “Not for Me,” as well as optionally write in your own note or tip. Meanwhile, you can surface those others have tried in a new activity feed, which also displays the percentage of Pinners who would recommend the Pin, based on the happy/sad feedback collected. This feed will point you to things like recipes that people have actually cooked, items they’ve bought, projects they’ve attempted, places they’ve visited, and more. Depending on region, the types Pins in this feed may vary. For example, those in the U.S. try more food and drink, fashion and DIY ideas, while U.K. users try more home décor, fashion and beauty ideas, the company found during early tests. Pinterest says the new feature is rolling out this week to iOS and Android and will arrive on the web in the weeks ahead.
|
null |
Ron Miller
| 2,016
| 11
| 17
| null |
Here are some more details of how Snap’s Spectacles work
|
Matthew Panzarino
| 2,016
| 11
| 10
|
Snap’s Spectacles are starting to , and . I’ve been able to glean some details about the way they work from early users. The images above and below come from the auction listing, which has already hit $553 as I write this. The Spectacles charge in the case and there are about 4 full charges stored in the case’s battery. You recharge the case with a cable attached to the wall. BI dug up a patent a couple of days ago that details the charging method, which consists of . This makes sense because true inductive charging requires a coil, which would require additional hardware and complexity in both the case and the glasses. You tap a button on top to start the snap, which lasts 10 seconds or (probably) until you tap again. There is an LED inside the right side of the Spectacles (behind the camera) that begins counting down the 10 seconds in a circular ring, showing you how much time you have left to snap. You can shoot around 100 ten-second clips using the Spectacles’ internal storage before you have to connect them to your phone and transfer them over. The Spectacles dump the pics automatically and drop into Snapchat’s Memories section. They feel like a pair of Ray Ban Wayfarers weight wise, so on the heavier side for sunglasses, which is to be expected given the camera unit embedded on the right side. The snaps themselves are extremely wide angle and have that “Point Of View” feel becoming more and more common in cameras and VR/AR systems. There is another whole discussion in how POV, which encourages eye contact (something that you get with Spectacles) is an enormous break in narrative technique from centuries of film and photography. Traditionally, eye contact with the camera was considered something to be avoided as it breaks immersion. With POV cameras, however, you’re placed into the narrative and have to come to grips with the fact that you’re a character. But I digress. Now that the Spectacles are getting out to people I’m sure we’ll see more about the way they’re used. It’s interesting that Snap is letting them get out to customers via a vending machine near its headquarters, rather than shipping them to traditional tech reviewers or press to fondle — but it makes a ton of sense to me. Interesting strategy for an interesting company. Here's what it's like to watch a Snap shot on Spectacles — Matthew Panzarino (@panzer) Ellen DeGeneres got an early peek at the vending machine and Spectacles on her Snapchat, so check that out at ‘Ellen’ on there. None of the people Snapchat wants to use these Spectacles are going to give 2 poops what a gadget pundit thinks. Can't wait to see the angst — Matthew Panzarino (@panzer)
|
How data science and rocket science will get humans to Mars
|
Kapil Kedar
| 2,016
| 11
| 19
|
In a recent to CNN, President Obama re-affirmed America’s commitment to sending a manned mission to Mars. Think your data science challenges are too complicated? Imagine the difficulties involved in mining data to understand the health impacts of an expedition to Mars. What happens to astronauts’ muscle tone or lung capacities after several years in space? How much weight can they safely lose? How much CO2 should be in the crew vehicle? How many sensors are needed to calculate joint flexibility in each individual space suit? When sending humans “where no one has gone before,” there are a multitude of variables to consider, and NASA is hard at work of a future Mission to Mars. Understanding these risks is critical, as they impact a number of decisions that need to be made when planning the journey — spanning everything from how potential crew members are evaluated to equipment engineering, mission logistics and the determination of needed fuel loads. The stakes are high, but NASA realized from the get-go that it needed to focus less on developing the perfect analytic model and more on building a data science that empowers decision-makers to use analytics to answer a multitude of continually changing questions. But you don’t have to be dealing with rocket science to learn from NASA’s analytic approach. Here are several key takeaways from NASA’s project that are useful for any organization about to embark — or that’s stuck — on a big data analytics initiative. Simply put, data science shouldn’t be as complicated as rocket science. (See what I did there?) Yes, analyzing big data has challenges, and yes, your approach may vary depending on what kinds of insights you hope to obtain, but there’s no need to make things more complex than the situation calls for. All too often, organizations end up spending endless cycles attempting to move data in order to analyze it when they should instead be focusing on bringing the analytics to the data. Big data, by definition, is very tough, if not impossible, to move around. This is why distributed storage and processing frameworks like Hadoop exist — data in the cloud is far more scalable than data in a silo. For the Mars project, there are so many levels of data to look at, ranging from health data collected from astronauts like who have completed previous space missions, to non-astronaut test studies, to studies done in simulated space environments like the (HERA) at Johnson Space Center in Houston. Getting all the data in one place is the critical first step. For this reason, NASA is using the platform developed by Lockheed Martin and several analytic partners, such as Alpine Data, to analyze data at its source. Because there’s no waiting to download data into a separate analytic environment to work with it, researchers can focus their time and energy on asking questions and getting the answers that will help them plan a mission to Mars. A successful rocket launch is only step one in a multi-year expedition to Mars. Based on past experiences, NASA expects to encounter and address numerous challenges along the way. The same holds true for data analytics projects. Simply deploying a model doesn’t mean the project is done. In fact, the most valuable analytics initiatives are those where models are continually refined and iterated on an ongoing basis. Like the scientific method, getting the most out of analytics requires experimentation, testing, learning from failures and testing again. NASA wants to be able to quickly query the large volumes of data at its disposal, then funnel insights back into new models capable of building on what came before. That’s why the data science process for this initiative resembles a “pendulum,” where the forward swing focuses on rapidly driving insights out to researchers and the backward swing focuses on measuring, evaluating results, refining the model and then swinging again. An ability to quickly and easily refine analytic models is especially valuable when your data sets aren’t perfect. (And really, is any data set perfect?) For NASA, the biggest data challenge is that the astronaut sample size is small — only 300 individuals have been accepted to NASA’s Astronaut Corps. Researchers have to mine the heck out of the data collected from this small sample and extrapolate. For example, based on how a 35-year-old female with a starting weight of 120 pounds responded to a five-month trip in space, how would a 32-year-old weighing 123 respond to two years? What about a 30-year-old weighing 118? Furthermore, since an astronaut has yet to step foot on the Red Planet, there’s no data about the health impacts of actually living on Mars (Matt Damon doesn’t count). But what can NASA learn from astronauts who have gone to the Moon, or spent a year in the International Space Station? What happens when data from test subjects living in simulated space environments is plugged into a predictive model? With analytic tools that support rapid model deployment and refinement, organizations can keep trying different ways to extract insight from the data they have to make better predictions, even when key information is missing. With the Mars Mission, NASA is not only putting billions of invested taxpayer dollars on the line, but also the lives of its astronauts, who risk their health and safety in the name of science and exploration. Like any consumer of analytics, NASA needs to be able to trust in the recommendations that are being generated, but this is hard to do if predictions are computed in a “black box” that only data science experts can manipulate or understand. For a project like this, empowering analytic consumers who aren’t necessarily data science PhDs (such as the health researchers, equipment engineers and others planning the mission) to actually build and launch queries and use the data on their own is key. This requires tight collaboration between business and IT stakeholders, modeling tools that are simple to use and modify and the ability to push insights to the people who need them. This is why NASA has chosen a collaborative analytic platform that includes tools that extend outputs directly into the systems and applications that are used by the scientists and decision-makers working on the Mars Mission. Large and complex data sets pose challenges for any organization about to embark on an analytics deployment. But NASA’s example of harnessing data to plan the most complicated of journeys — an expedition to Mars — proves that the challenges are not insurmountable. With the right tools and, most importantly, a consistent and well-planned approach, data science doesn’t have to be as daunting as rocket science.
|
This crowdfunded watch looks like the Chrysler Building but was designed in Honolulu
|
John Biggs
| 2,016
| 11
| 19
|
The is a new crowdfunded piece by horologist Mark Carson and designer Richard Paige. Based on Art Deco skyscraper decoration – think the Empire State Building’s handsome arches or the swooping buttresses of Gotham city – this watch is interesting for a few reasons. First, this team has made some nice watches before. Their dual face watch was quite popular with the watch bloggers and forums and this one, while a bit simpler, is still quite clever. Built around a the massive Swiss ETA/Unitas 6497 movement, essentially a hand wound pocket watch movement. The watch is encased in a Chrysler Building shaped steel frame with sapphire crystal with exhibition back. Carson is based in Honolulu but has 150 of the pieces ready to ship and should be able to get them out to backers by December 16. There are only a few left at $925 and they come on leather straps of your choosing. Again, as the resident TC watch hound I consider these guys to be quite unique. It’s not often you get an American-designed limited edition watch on the Kickstarter market and it’s even rarer to find them from truly established watch folks. These could be a good Christmas present or, considering the original name of these pieces was the Crash of ’29, a prescient piece of memorabilia as we slide into January.
|
Watch a fully self-driving Tesla navigate intersections and city streets
|
Darrell Etherington
| 2,016
| 11
| 19
|
[vimeo 192179727 w=680 h=383] Tesla revealed it would be building all vehicles complete with all the hardware onboard they need to achieve full self-driving in the future, and also published a short demo video of the prototype software in action at the time. Now, a new video shows a self-driving Tesla making its way through a city neighbourhood complete with a lot of traffic and quite a few intersections. The video provides a driver’s eye view of the trip, including a clear look at how the human behind the steering wheel (who’s there only for legal compliance reasons) never once touches either the wheel or the brake and gas pedals. The video also shows three separate views of what the Tesla optical cameras are seeing at all times during the drive, including colored overlays showing what the onboard computer is detecting in terms of objects and lane markers. This new look at the fully self-driving version of Autopilot in action is definitely impressive, and helps boost confidence that Tesla might even be able to meet founder and CEO Elon Musk’s goal of doing a . There’s also a Benny Hill cut of the video, in case the backing track on the longer version is a bit too heavy for your tastes. [vimeo 192179726 w=680 h=383]
|
After decades of dreaming, indoor location’s time has come
|
Arif Janmohamed
| 2,016
| 11
| 19
|
The precise information you need, delivered exactly when and where you need it. Personalized engagement and experiences. Easy step-by-step navigation. These are just some of the things you get with indoor location-based services. Similar functions have existed outdoors for more than a decade thanks to GPS and apps like Google Maps. However, indoor location services have seen much slower adoption. But that’s started to change, and in 2017 we’ll see it shift into the mainstream. Indoor location’s time has finally come, and here’s why… The first catalyst for this shift is changes in our smartphones. and , both of which make Bluetooth Low Energy (BLE) a standard on virtually every smart device. This means that every smartphone is now “location aware,” allowing stores, hospitals, schools, corporations, museums and more to deliver highly personalized services to anyone who walks in their doors. In my view, this is as significant as when Intel added Wi-Fi to the Centrino processor in 2003. Once it was ubiquitous on every laptop, the technology took off like wildfire. Going into 2017, we are at a very similar inflection point with BLE. The second catalyst, which is actually a critical enabler for the first, is the introduction of new wireless infrastructure that takes advantage of BLE and makes it easy and cost-effective to deploy at scale. The single biggest thing that has held back BLE deployments to date, for example, has been physical beacons — hockey puck-sized devices that need to be placed every 20 feet or so to detect BLE users and then engage with them. They are expensive and time-consuming to deploy, and a , especially in environments that change frequently. By virtualizing this beaconing function, new wireless infrastructure has removed this obstacle. Just last month, for example, Cisco — the 800-pound gorilla in the industry — a new product using this approach. The second biggest infrastructure challenge to BLE deployment has been location accuracy. To use BLE for strategic functions like step-by-step navigation, asset tracking and push notifications, indoor location accuracy needs to be close to the one-meter range. Advances in machine learning, coupled with rolling BLE into Wi-Fi systems, have recently made this possible. The final catalyst is motivation. Mobile users expect to be delighted by everywhere they go — inside as well as out. And companies want to engage more closely with users on their premises to deliver amazing mobile experiences. Hotels want to automatically check-in guests the minute they walk in a front door and give them step-by-step directions to their room or a nearby restaurant. Hospitals want to help people easily find their way to the lab or the cafeteria while also easily locating valuable equipment such as wheelchairs and infusion pumps. Retail stores want to notify shoppers of sales within specific departments and make it easy to find available staff for assistance. And in this age of data, companies expect the analytics that come with the above: Where are my customers spending their time? How long are they there? What are they doing in each location? This information is critical to improving the bottom line, while offering services that reflect actual consumer needs. None of this potential is lost on private investors who have been placing substantial bets on location startups this year. , a targeted-ad startup, snagged $25 million in a Series B offering. Advertiser-location intel company received $25 million in a Series D. received a $20 million Series C. , which uses machine learning to deliver high-accuracy location and business-critical Wi-Fi services, recently announced a $28 million Series B round (disclosure: my firm is among Mist’s investors). And , which makes location-savvy mobile apps, drew an additional $22 million recently, for a total of $90 million in funding. I’m excited by the possibilities for these and other companies in the location ecosystem. If they can deliver on their potential, we’ll all be able to get the personalized experiences we have been promised for years.
|
Snapchat Spectacle vending bot lands at the Rose Bowl Stadium in Pasadena
|
Darrell Etherington
| 2,016
| 11
| 19
|
Another day, another Spectacles vending machine location. This is the fifth place the Snapbot has appeared, to sell its first-person video capturing sunglasses to droves of willing Snapchat fans (or at least people hoping to turn a tidy profit by reselling them on eBay). Today’s spot is in Pasadena, at the Rose Bowl Stadium where there’s a USC/UCLA game later today. The previous locations include Venice Beach, Big Sur, Tulsa, and Santa Monica. The random locations suggest little in the way of rhyme or reasons beyond keeping things unpredictable, which is definitely helping with the hype cycle. The lack of a predictable pattern also means that demand at any given location isn’t overly artificial – it’s people who are able to get there with almost no advance warning, basically. This time is different, in that the Snapbot has appeared near a spot where there’s a big event happening soon. The USC/UCLA game will draw a huge crowd, and it’ll also draw a large number of tailgaters before the game itself. The idea that Snapchat could start tying Snapbot pop-ups more specifically to special events and locations likely to draw a big crowd is an interesting one. They’re definitely a good fit for this kind of thing, since people tend to share more when they’re at an event like a festival or a game. This could also be just another variant in its loose trend of putting the bots at interesting places, however. If you’re catching the game, let us know what the lineup looks like. Snap has been selling out quickly even in remote locations, so it’ll be interesting to see how quickly this one goes or if Snapchat refills them once they sell out.
|
Behind The Scenes with the tech powering Cirque Du Soleil’s newest show, Luzia
|
Greg Kumparak
| 2,016
| 11
| 19
|
When the folks at Cirque Du Soleil invited us to check out some of the technology behind its latest show, , I just couldn’t say no. I try to get to every Cirque show that comes to the Bay Area, so this was like getting a glimpse into Santa’s workshop. They’ve built all kinds of incredible stuff for this show, from a monstrous treadmill that runs the length of the stage, to a dynamic waterfall wherein each drop is independently controlled to rain images of flowers and birds down from the bigtop, to a dress covered in flowers that subtly bloom to life. It all comes together to help elevate the show without overpowering its human element; it compliments the performers, but doesn’t distract from the absurd, superhuman-level stuff they’re pulling off night after night. I asked if we could bring a camera along, and they were open to it — so join us for this rare behind the scenes look!
|
Zuckerberg reveals plans to address misinformation on Facebook
|
Kate Conger
| 2,016
| 11
| 19
|
problem persists, CEO acknowledged last night. “We’ve made significant progress, but there is more work to be done,” Zuckerberg wrote, outlining several ways to address what he called a technically and philosophically complicated problem. He proposed stronger machine learning to detect misinformation, easier user reporting and content warnings for fake stories, while noting that Facebook has already to eliminate fake news sites from its ad program. The firestorm over misinformation on Facebook began with a particularly outrageous headline: “FBI Agent Suspected in Hillary Email Leaks Found Dead.” The false story led to accusations that Facebook had tipped the election in Donald Trump’s favor by turning a blind eye to the flood of fake stories trending on its platform. The story, which ran just days before the election on a site for a made-up publication called Denver Guardian, suggests that Clinton plotted the murders of an imaginary agent and his imaginary wife, then tried to cover it up as an act of domestic violence. It was shared more than 568,000 times. The Denver Guardian story caused a crisis at Facebook, and it hasn’t gone away. Last night, the
story appeared yet again in a friend’s newsfeed. “BREAKING,” the post blared. “FBI AGENT & HIS WIFE FOUND DEAD After Being ACCUSED of LEAKING HILLARY’s EMAILS.” This time, the story was hosted by a site called Viral Liberty. Beneath the headline is a button encouraging Facebook users to share the story, and according to Facebook’s own , it’s been shared 127,680 times. Facebook isn’t alone. Google and Twitter grapple with similar problems and have mistakenly allowed fake stories to rise to prominence as well. And although stories about the rise of fake news online have focused primarily on pro-Trump propaganda, the sharing-without-reading epidemic exists in liberal circles too — several of my Facebook friends recently shared an article by the ‘s satirist Andy Borowitz titled “ ” as if it were fact, celebrating in their posts that Trump may not dismantle the Affordable Care Act after all his campaign promises to the contrary. But, as the hub where 44 percent of Americans read their news, Facebook bears a unique responsibility to address the problem. According to former Facebook employees and contractors, the company struggles with fake news because its culture prioritizes engineering over everything else and because it failed to build its news apparatus to recognize and prioritize reliable sources. Facebook’s media troubles began this spring, when a contractor on its Trending Topics team told that the site was biased against conservative media outlets. To escape allegations of bias, who vetted and wrote Trending Topics blurbs and turned the feature over to an algorithm, which quickly began from sites designed to churn out incendiary election stories and convert them into quick cash. It’s not a surprise that Trending Topics went so wrong, so quickly — according to Adam Schrader, a former writer for Trending Topics, the tool pulled its hashtagged titles from Wikipedia, a source with its own . “The topics would pop up into the review tool by name, with no description. It was generated from a Wikipedia topic ID, essentially. If a Wikipedia topic was frequently discussed in the news or Facebook, it would pop up into the review tool,” Schrader explained. From there, he and the other Trending Topics writers would scan through news stories and Facebook posts to determine why the topic was trending. Part of the job was to determine whether the story was true — in Facebook’s jargon, to determine whether a “real world event” had occurred. If the story was real, the writer would then draft a short description and choose an article to feature. If the topic didn’t have a Wikipedia page yet, the writers had the ability to override the tool and write their own title for the post. Human intervention was necessary at several steps of the process — and it’s easy to see how Trending Topics broke down when humans were removed from the system. Without a journalist to determine whether a “real world event” had occurred and to choose a reputable news story to feature in the Topic, Facebook’s algorithm is barely more than a Wikipedia-scraping bot, susceptible to exploitation by fake news sites. But the idea of using editorial judgement made Facebook executives uncomfortable, and ultimately Schrader and his co-workers lost their jobs. “[Facebook] and Google and everyone else have been hiding behind mathematics. They’re allergic to becoming a media company. They don’t want to deal with it,” former Facebook product manager and author of Antonio Garcia-Martinez told TechCrunch. “An engineering-first culture is completely antithetical to a media company.” Of course, Facebook doesn’t want to be a media company. Facebook would say it’s a technology company, with no editorial voice. Now that the Trending editors are gone, the only content Facebook produces is code. But Facebook is a media company, Garcia-Martinez and Schrader argue. “Facebook, whether it says it is or it isn’t, is a media company. They have an obligation to provide legit information,” Schrader told me. “They should take actions that make their product cleaner and better for people who use Facebook as a news consumption tool.” Garcia-Martinez agreed. “The has a front page editor, who arranges the front page. That’s what readers read every day — what the front page editor chooses for them. Now Mark Zuckerberg is the front page editor of every newspaper in the world. He has the job but he doesn’t want it,” he said. Zuckerberg is resistant to this role, writing last night that he preferred to leave complex decisions about the accuracy of Facebook content in the hands of his users. “We do not want to be arbiters of truth ourselves, but instead rely on our community and trusted third parties,” he wrote. “We have relied on our community to help us understand what is fake and what is not. Anyone on Facebook can report any link as false, and we use signals from those reports along with a number of others — like people sharing links to myth-busting sites such as Snopes — to understand which stories we can confidently classify as misinformation.” However, Facebook’s reliance on crowd-sourced truth from its users and from sites like Wikipedia will only take the company halfway to the truth. Zuckerberg also acknowledges that Facebook can and should do more. “There’s definitely things Facebook could do to, if not solve the problem, at least mitigate it,” Garcia-Martinez said, highlighting his former work on ad quality and the Facebook uses to remove images and posts that violate its community guidelines. To cut back on misinformation, he explains, “You could effectively change distribution at the algorithmic level so they don’t get the engagement that they do.” This kind of technical solution is most likely to get traction in Facebook’s engineering-first culture, and Zuckerberg says the work is already underway. “The most important thing we can do is improve our ability to classify misinformation. This means better technical systems to detect what people will flag as false before they do it themselves,” he wrote. This kind of algorithmic tweaking is already popular at Google and other major companies as a way to moderate content. But, in pursuing a strictly technical response, Facebook risks becoming an opaque censor. Legitimate content can vanish into the void, and when users protest, the only response they’re likely to get is, “ .” Zuckerberg is rightfully wary of this. “We need to be careful not to discourage sharing of opinions or to mistakenly restrict accurate content,” he said. Mike Caulfield, the director of blended and networked learning at Washington State University Vancouver, has critiqued Facebook’s misinformation problem. He writes that sharing fake news on Facebook isn’t a passive act — rather, it trains us to believe the things we share are true. “Early Facebook trained you to remember birthdays and share photos, and to some extent this trained you to be a better person, or in any case the sort of person you desired to be,” , adding: The process that Facebook currently encourages, on the other hand, of looking at these short cards of news stories and forcing you to immediately decide whether to support or not support them trains people to be extremists. It takes a moment of ambivalence or nuance, and by design pushes the reader to go deeper into their support for whatever theory or argument they are staring at. When you consider that people are being trained in this way by Facebook for hours each day, that should scare the living daylights out of you. When users look at articles in their News Feed today, Caulfield notes, they see prompts encouraging them to Like, Share, Comment — but nothing suggesting that they Read. Caulfield suggests that Facebook place more emphasis on the domain name of the news source, rather than solely focusing on the name of the friend who shares the story. Facebook could also improve by driving readers to actually engage with the stories rather than simply reacting to them without reading, but as Caulfield notes, Facebook’s business model is all about keeping you locked into News Feed and not exiting to other sites. Caulfield’s suggestions for an overhaul of the way articles appear in News Feed are powerful, but Facebook is more likely to make small tweaks than major changes. A compromise might be to label or flag fake news as such when it appears in the News Feed, and Zuckerberg says this is a strategy Facebook is considering. “We are exploring labeling stories that have been flagged as false by third parties or our community, and showing warnings when people read or share them,” he said. It’s a strategy that sources tell me is being considered not just at Facebook but at other social networks, but risk-averse tech giants are hesitant to slap a “FAKE” label on a news story. What if they get it wrong? And what about stories like Borowitz’s satire — should the story be called out as false, or merely a joke? And what if a news story from a legitimate publisher turns out to contain inaccuracies? Facebook, Google, Twitter, and others will be painted into a corner, forced to decide what percentage of the information in a story can be false before it’s blocked, downgraded, or marked with a warning label. Like the fight against spam, clickbait, and other undesirable content, the war against misinformation on platforms like Google and Facebook is a game of wack-a-mole. But both companies have built their own interfaces for news — Accelerated Mobile Pages and Instant Articles — and they could proactively counter fake stories in those spaces. AMP and Instant Articles are open platforms, so fake news publishers are welcome to join and distribute their content. But the companies’ control over these spaces gives them an opportunity to detect fake news early. Google and Facebook both have a unique opportunity to fact-check within AMP and Instant Articles — they could place annotations over certain parts of a news story in the style of to point out inaccuracies, or include links to other articles offering counterpoints and fact-checks. Zuckerberg wasn’t clear about what third-party verification of the news on Facebook would look like, saying only, “There are many respected fact checking organizations and, while we have reached out to some, we plan to learn from many more.” Bringing third-party vetting back into the picture means a return to the kind of human oversight Facebook had in its Trending Topics team. Although Facebook has made clear it wants to leave complex decisions up to its algorithms, the plummeting quality of Trending Topics makes it clear that the algorithm isn’t ready yet. “I don’t think Trending ever had a problem with fake news or biases necessarily, before the Gizmodo article or after. All the problems were after the team was let go,” Schrader said, noting that Facebook intended to incorporate machine learning into Trending Topics but needed human input to guide and train the algorithm. Engineers working on machine learning have told me they estimate it would take a dedicated team more than a year to train an algorithm to properly do the work Facebook is attempting with Trending Topics. Zuckerberg did acknowledge that perhaps Facebook can learn something from journalists like Schrader after all. “We will continue to work with journalists and others in the news industry to get their input, in particular, to better understand their fact checking systems and learn from them,” he said. But the media certainly isn’t perfect. Sometimes we get our facts wrong, and the results can range from comical to disastrous. In 2004, the issued questioning its own reporting on several factually-inaccurate stories that spurred the war in Iraq. As journalists sometimes make mistakes, so will Facebook. And when that happens, Facebook should address the errors. “In a small back door sort of way, it will adopt some of the protocols of a media company,” Garcia-Martinez says of Facebook. One suggestion: “Get a public editor like the .” The public editor serves as a liaison between a paper and its readers, and provides answers about the reporting and what could have been done better. In his late-night Facebook posts, Zuckerberg has already somewhat assumed this role. But an individual with more independence could help Facebook learn and grow. “They are going to get a lot better about this business of editorship,” Garcia-Martinez predicts. “When the stakes are American democracy, saying, ‘We’re not a media company,’ is not good enough.”
|
Gillmor Gang: Real News
|
Steve Gillmor
| 2,016
| 11
| 19
|
The Gillmor Gang — John Borthwick, Keith Teare, Frank Radice, Kevin Marks, and Steve Gillmor. Recorded live Friday, October 29, 2016. Mainstream media meets streaming media as we await the transition to the new administration. The Gang tackles the fake news pandemic with a requiem for truthiness scheduled for January 20th. @stevegillmor, @borthwick, @kevinmarks, @fradice, @kteare Produced and directed by Tina Chase Gillmor @tinagillmor
|
Hitman creative director looks back at Hitman’s first season and games as a service
|
Romain Dillet
| 2,016
| 11
| 19
|
is a weird video game franchise. You incarnate an assassin-for-hire and you have to kill various targets in different cities around the world. But you get to decide how you want to do it. While many are putting a lot of emphasis on cut scenes, story lines and closed environments, game developer is taking an opposite approach by going all-in on sandbox gameplay. But that wasn’t always the case. While the game series is 16 years old today, it has been a bumpy ride. In 2012, provided a much more linear gameplay. In other words, the 2012 episode looked like countless of other games out there. That’s why IO Interactive took a risky yet ambitious bet. The Hitman franchise needed a reboot, and the company wanted to go back to the original game’s core features. Hitman is a sandbox game with huge levels and countless of possibilities to kill your targets. It relies on your creativity and sense of observation, but it can also feel like a daunting game. When the company announced the game last year, it also said that the game would be released in six episodes over the course of 2016. Many fans were worried, thinking that a game like Hitman didn’t make any sense in an episodic format. I’ve played the game for dozens of hours, and it was by far my favorite Hitman game. Each level was deep enough that you could easily play for 10 or 20 hours trying all sorts of disguises and contracts. I interviewed IO Interactive creative director Christian Elverdam right before the release of the last episode. It was a good opportunity to look back at the Hitman reboot. This interview was slightly edited for brevity and clarity. I think we’re quite happy right now. It’s been quite a journey building all these levels and shipping them out. Now, looking at everything together, it’s very nice. It’s also been nice to follow people talking about the episodes together and digging into all the details of each individual level. Also, we managed to bring back sandbox-typed gameplay after ten years and people liked it. That was obviously really exciting at the beginning of the season. How would people receive the game? Would they get it? Would they like it? I think there are more gamers now. And also more generally we’re all getting older, the average gamer age is just getting higher because all of us are now gamers. 30 years from now, we’ll have retired people who play games, which is still uncommon right now. I just think, as a medium, we’re maturing. More specifically, for Hitman, what’s different this time around is that the episodic format gives people more time to get used to the depth of the game, especially for people who were not into a game like . And I think that’s new for us, even when looking at as well. We’ve managed to make more people understand how to approach a deep and complex sandbox game. I think they will. If you spend any time reading articles or watching videos on the web for the last seven months, you’ll see that a lot of people are talking about the depth of the game. It makes it much easier for people who are jumping in. I also think that part of the people who jump in right now are people who might know Hitman really well but they want the full package so they can play it at their own pace. And we said that it’s fine — if you want to wait, you should wait. We don’t want to change that mindset. So I think people who jump in now have a good chance of understanding just how much is in there. We’ll also have to see if there���s a difference in behavior. The risk obviously is that they might not get the depth of the game. Based on how many people have talked about the game already, I’m pretty confident that they will. We feel like Hitman is a success. We can’t talk about that too much specifically. We’re pretty happy. Again, I can’t be too specific. But in general, it’s been going on the entire season. One of the advantages of building a game that is live is that there are many entry points to jump into the game. Also we’re still continuing to release elusive targets and escalation contracts. I think we’ve managed to create a game with many starting points and I can also see that in how people buy it. You can now release patches for console games, it has become much easier. A few years ago, when we decided to work on this version of the game, part of what made it possible is that Microsoft and Sony changed their policies so that developing a game for a console would feel more like developing a game for the PC. That trend in itself will lead a lot people to think like us. Based on our conversations with other people in the industry, a lot of people are looking at what we’re doing and think it’s fascinating. I think there could be a lot of people who want to go this way. I still think it needs to fit with the video game world at some level. For us, the concept of Agent 47, an assassin who is getting his assignments ever so often — you’re going to Paris, you’re going to Morocco, you’re going to Japan… That felt like a good fit with the world of Agent 47. You get these missions that pop up. “Oh, by the way, there’s a target in Italy for the next 48 hours, you need to hurry.” All that stuff plays a part when it comes to building a game like that. Our game is a single player game. So I think we’re a proof that you can do this with a single player game. I think it’s also really tied with the video game universe. Does it help the universe? When you get a notification that there’s an elusive target, a part of you feels a little bit like an agent with an assignment. And that’s why it clicks with you. I don’t know if any game could do it, but I think many games could probably do it. A lot of people are asking about that. I take that as a sign that people really like what we’re doing. Going back to when we announced the game, I talked about it at E3 a few years ago. We had this ambition to build a world that would expand. And we talked about the fact that we would like to tell a story that could span over multiple seasons of Hitman. So our approach, our thoughts and our overall vision for the game is something that lasts more than a season obviously. Hitman is like what you’d expect from your favorite TV shows. It closes some story arc at the end of the season, and then it also opens some new doors for the future. I can’t really be super specific about it but we’re definitely thinking about something that will continue. Obviously, we need time to build a season. But at the moment, we’re wrapping up season one and we’re looking back at the season. We’re fortunate as we’re looking back at a season during which a lot of things worked. Now, it’s more about thinking how we can improve it. And then, at some point we’ll figure out what’s next. We had to fundamentally rethink a lot of things. We had big ambitions and such a small amount of time. Some things are planned a long time in advance. We know where the story is going to go, we know the locations as well because we need to start building the geometry and all that. But there’s a difference between knowing the locations and the way we build it. With the final episode in Japan, Hokkaido was heavily influenced by having shipped both Paris and Sapienza. Seeing that the sandbox was well-received, it gave us some confidence to shape up the Colorado and Hokkaido levels. There are many things that we’ve been playing around with, and I don’t think we’d have played around with them if we had released everything at once. There’s nothing more honest than watching people playing your game. That’s the mindset we’ve been in for the last several months. Previously, we shipped a game and then you could see what you did for two weeks and understand how people received it. This time, it’s been a longer period. We can tweak the game. I’ve been there for almost nine years now. We shipped a game over a period of seven months. The way we did that was by actually not working crazy hours all the time. Some years ago, the industry was at a different place. People worked crazy hours before releasing a game, then shipped a game, went on vacation and took a break. When you do that, you start doing a lot of mistakes and a lot of errors. We had seven months ahead of us during which we needed to execute with almost no error, otherwise we would be delayed. Also, as human beings, you can’t work like crazy for that long. We’ve worked with some pressure, but it hasn’t been crazy all the time. That’s definitely something that feels really good. It has become more professional at IO and in the industry in general as well.
|
ShotSpotter CEO Ralph Clark talks about the future of city surveillance
|
John Biggs
| 2,016
| 11
| 19
|
A guy like Ralph Clark has a tough job: if all goes according to his business plan he’ll put himself out of business. He is the CEO of ShotSpotter, a surveillance system for cities that triangulates gunfire in 90 cities. I asked him if he’d ever have to shut down. “You know, sadly, in the US that won’t be the case,” he said. The company for being seemingly ineffective in many cities. The problem? Cops visit the places where the system heard shots but they find nothing. That’s exactly how it should be, said Clark. The goal isn’t to stop gunfire but to make it clear to the residents of that area that the police are listening and that they are ready to investigate when something dangerous happens. While while other Ultimately, however, Clark believes that only a small number of individuals are responsible for most of a city’s gunfire and any tools available to get those folks off the street are important. Clark has some fascinating ideas about preventing poaching using his technology and here’s hoping that some day his devices won’t hear a single shot anywhere. You can subscribe in or and .
|
The Smart Garden 9 is a self-contained kitchen garden for your “herbs” and vegetables
|
John Biggs
| 2,016
| 11
| 19
|
When we last met Click & Grow they were designed for indoor gardening. Their new product, called the Smart Garden, is a self-contained kit pod for growing fruits, vegetables, and “herbs” that costs $129 on The system is quite simple. You fill the water tank, place in small pods of soil, nutrients, and seeds and then turn it on. It’s automatically lit by a set of LEDs and all you have to do is raise the lights when the plants get too high. The entry level gets you all lettuce pods but there are also seedless pods so you can grow anything you want. The kit can grow nine plants and automatically waters your pods as necessary. Mattias Lepp, founder of Click & Grow, was an orchestra conductor who went into IT. Telia acquired his first IT company in Estonia and he decided to return to his long-term interest in plant cultivation. “We have over 300,000 active users for our products thus far,” he said. “The main volume of users today is made up by our second generation product users, the Smart Herb Garden, but we’re seeing a lot of traction happening already for the larger wall farms directed at consumers as well.” “The Smart Garden 9 got it’s start from surveys we made among our clients,” he said. “More than 80% indicated an interest in having a bigger garden than our previous one. The bigger bulk of the 9-hole product specific wishes came from the backers of our last project on Kicsktarter. So we decided to give people what they want and also improved some key features on the new garden: we featured high-end LED lights that offer plants the perfect spectrum to induce photosynthesis while focusing the lenses in a way that there’s much less light pollution coming from the LEDs.” The company raised $4.1 million from a number of investors including Jaan Tallinn and Ruchi Sanghvi. They’ve also received some government funding to make the smart soil product. The Smart Garden 9 ships next July but I saw an early version and it looks like what Apple would make if it really liked tiny peppers. I approve.
|
The tech that will feed the world
|
Joseph Byrum
| 2,016
| 11
| 19
|
Twenty years from now, the most important tool for putting food on your table won’t be a harvester, combine or a plow. It will be a piece of software. Agriculture is in the process of transitioning into a fully high-tech enterprise. This is a long-overdue revolution in the way things have been done for centuries. To put it in perspective, if we keep doing farming the old-fashioned way, two billion more people will go hungry by the year 2050. World population growth is driving the urgent need for a radical boost in farm productivity. Incremental advances in output simply won’t cut it. The challenge in 2050 will be unprecedented — equivalent to growing enough food to feed everyone who is alive today, plus everyone who was alive in 1920. Simply put, the techniques that got us through the 20 Simply put, the techniques that got us through the 20 century won’t get us very far into the 21 . And whatever solutions we come up with will to solve this dilemma will also have to take into account that land and water continue to be scarce resources, and environmental sustainability remains paramount. Fortunately, high-tech gadgets are spreading to farms across the country and nudging crop productivity upward. While self-driving cars are having a tough time achieving acceptance on America’s highways, self-driving combines and tractors are an increasingly commonplace sight in America’s wheat and corn fields. Tesla s last year, compared to John Deere, in the field. Drones and satellites are giving farmers an unprecedented overview of the health of their crops while arrays of ground-level sensors provide real-time data on soil and climate. These systems provide an early warning of the presence of invasive insects or other problems that may threaten crop survival. A soil map created using agricultural drones and PrecisionHawks’ DataMapper. A well-informed farmer is able to act quickly to fix problems before they become serious. For instance, nitrogen sensors may report an excess in one part of a field, but in another small patch there might be a shortage. This allows the farmer to adjust an advanced fertilizer delivery system to ensure nutrients are dispensed in precisely the amount needed—not too much, not too little. Precision prevents waste, which saves money, but it also is good for the environment. Growing plants crave and absorb nitrogen, but farmers without the latest equipment sometimes overdo it by providing extra fertilizer “just in case” the crops need more. Unfortunately, leftover nitrogen that isn’t absorbed by plants has a tendency to seep into groundwater where, in excess, it can be harmful to fish. So the potential benefits to efficiency and to the environment from these high-tech gadgets are tremendous, but they represent a single piece of a much more complex puzzle. What gadgets do is unlock an unprecedented level of monitoring and data collection capabilities. But the heart of 21st-century agriculture will be the processing of these data. The killer app of tomorrow’s agriculture is information harvesting. Farmers face an array of questions when it comes to growing. What crop should be planted, and where, and when? How much water is needed? How much fertilizer is needed? The amounts of water and fertilizer will vary from field to field, and within individual fields. The amounts will also change by the day or sometimes by the hour. There are thousands of complex, interrelated variables involved in the process. ISRAEL – OCTOBER 03: Drip irrigation (Photo by James L. Stanfield/National Geographic/Getty Images) Back in the days when complex math could only be performed with a slide rule and chalkboard, we simply didn’t have the computing capacity required to answer complex growing questions to determine the best course of action with anything more than educated guesses. Now computing capacity is cheap, and it’s possible to model all possible choices and their potential outcomes. A smartphone with Google Maps, for example, can evaluate every path from point A to point B to decide, based on the current traffic conditions, which will likely be the shortest or fastest route. Simulation and modeling also help from getting lost when it comes to growing crops. At the most basic level, plants need sunlight, water and nutrients at levels that vary during various stages of growth. It sounds simple, but at scale, optimizing each factor has a huge payoff. The United States annually produces 2.3 billion bushels of wheat across 50 million acres of farmland. A one percent boost in productivity flour every year. We can do much better than a one percent increase by taking full advantage of the power of data analytics. Even before the farm’s first seed is planted in the ground, data analytics can be used to optimize the breeding of plant varieties to maximize the genetic potential for a variety of growing conditions around the country and around the world. Farmers in California might need a seed that tolerates drought, while a Midwestern farmer might want a seed with extra resistance to a particular plant disease. Prospera’s crop monitoring system uses computer vision and artificial intelligence to help farmers improve their yields. Data analytics can also help on the customer end, matching a particular farmer’s needs to the seed that will perform best in that farmer’s field under the expected harvest conditions for this year, not last year. When it comes time to plant that seed, data analytics can process vast amounts of historical data on crop growth, soil condition, weather and other key factors to ensure that individual plant conditions are optimized for maximum growth. After the harvest, data analytics assist with the logistics of distribution and sales of the crops. The information harvest must be considered as a complete system that optimizes every step of the process from the breeding of seed varieties to the placement of food products on store shelves. The idea is to improve the decisions made in agriculture, from those made by farmers and their suppliers to equipment makers and ultimately to consumers. Making the best possible choices at every stage of the food production process will help maximize productivity and bring us closer to meeting the goal of growing more food with fewer resources. This is the level of effort that will be needed to achieve the crop growth needed to feed the world in 2050. This is a task bigger than any one company or individual. It’s a challenge that even extends beyond the traditional agricultural community. Food security affects everyone, and solving the problem will require an infusion of talent from the worlds of mathematics and software engineering to build the innovative systems agriculture needs to reach the next level of productivity. What this all means is that many of tomorrow’s greatest tech opportunities will be found not be in Silicon Valley, but in the Midwest.
|
Technology as a force for division — and unification — in politics
|
Richard Muirhead
| 2,016
| 11
| 26
|
The functions of government outlined in the U.S. Constitution can be broadly broken down into the promotion of unity, justice, peace, defense and welfare. Nearly 230 years later, in an age when software rules the world, these principles remain constant — but the way in which the average citizen experiences and interacts with government and politics has changed immeasurably. Around the world, government services have been increasingly going online. In the U.K., a began in 2013, pushing through services that let citizens do things like claim certain welfare benefits, pay taxes and register to vote online. Indeed, the U.K. was ranked at the top of the for 2016. The report found that “countries in all regions are increasingly embracing innovation and utilizing ICTs to deliver services, increase transparency and engage people in decision-making processes.” Despite the U.S. not featuring in the top 10 of the E-Government Development Index, Barack Obama’s presidency has been underpinned by a real push toward the principles of e-government, where information and services are made available online as far as is possible. So the idea of government-as-a-service is nothing new — the idea that we can report issues as minor as or as major as using online tools are familiar to us all. But while the increase in digital government services has acted as a catalyst for people to become more socially responsible and empowered citizens, technology is changing how we act in a political sense — and not necessarily for the better. In terms of the news we see and read, some key things have changed over the years. Whereas printed publications still have editors who dictate what the audience sees, with online news, things are very different. Yes they still have editors, but news websites are pumped full of content that we arrive at in all sorts of ways — through social media, search, via an app — and the increasing personalization of this news means that we tend to see more about the things and people in which we are interested. Whether this has to do with sports, entertainment or politics, algorithms that present us with content are — unwittingly — skewing our compass. That “echo chambers” exist within many online communities is something that we have suspected for a long time — and now have of this, too. A study of Facebook users found a high degree of polarization within the social network, with users tending to interact most frequently with the people and narratives they agreed with — creating an echo chamber. Facebook also removed another human element from its platform recently. Staff tasked with curating news stories for the Trending Stories section — and there were some well-publicized teething problems. The story chosen to illustrate why newscaster Megyn Kelly was trending — — shows that algorithms aren’t necessarily guaranteed to be free of political bias, even if that bias is completely unintentional. Given that so many people now use Facebook as their , such political bias can further increase the polarization within the site, with this effect carried through into the real world, too. As Plato said, “those who tell the stories rule society.” With the populist political movements such as the anti-EU “Leave” group in the U.K. and Donald Trump’s camp being increasingly flexible with the truth, outrageous claims are hitting the headlines all the time, and are being used as justification for more and more extreme political views. Politicians from all sides use the web to push their own version of the story, and frequently it is not so important to be seen as honest as it is to be pushing a populist message that fits in with a group’s existing world view — however untrue it might be. And while some sections of the media aim to call out others they , this process often descends into mudslinging that makes both sides look bad. Even though we have reached the age of , algorithms can help us redress the balance. In October this year, Google announced that it was to the Google News service that would highlight articles that had been rigorously fact-checked. Alexa, Amazon’s virtual assistant, was the recent presidential campaign. Sites such as kept tabs on statements made by the candidates, ranking their relative level of truthfulness. But while these features and services are useful, they are unlikely to change the minds of those who have already been exposed to the echo-chamber effect for many years. With discourse becoming increasingly polarized, it becomes ever harder — and less excusable — for citizens to be apathetic toward politics. To go back to Plato, “the price of apathy towards public affairs is to be ruled by evil men.” Fortunately for us, it is now becoming easier to become politically active thanks to digital services. Rock the Vote teamed up with theSkimm to create an , and there are a number of websites that can help you decide based on how much you agree with key policies. Tinder’s was a particularly interesting — if somewhat gimmicky — innovation. goes one step further, not just matching your values to political parties, but crowdfunded individual causes, as well. But while governments and private enterprises need to continue to push to improve the lives of their citizens through the provision of digital services, we must be wary of the power that technology has over how we act politically. With much content designed to win clicks rather than promote genuine debate, and many platforms accentuating biased opinions, fact-checking algorithms could be our salvation. Whether they can completely counteract the echo-chamber effect, though, remains to be seen.
|
So you think you elected an autocrat
|
Jon Evans
| 2,016
| 11
| 19
|
Welp. I’m sorry to hear that. Whether you are Venezuelan, Zimbabwean, Russian, Turkish, or American; whether your democratic process has been fatally compromised, or your fellow citizens have knowingly voted in a monster; indeed, even whether you’re right or wrong — I feel your pain. But for what it’s worth, some technology, if used correctly, is still on your side. Here’s what to do: Approach the situation as an engineer would. If the worst-case scenario is cataclysmic, then you need to mitigate that risk, even if it’s low. Fifty years ago, notoriously liberal Afghanistan was full of hippie tourists, while hordes of Italian sunbathers flocked to Somalia’s beaches. Thirty years ago, Yugoslavia was the most cultured and Westernized nation in all of Eastern Europe, while Zimbabwe was the breadbasket and great hope of southern Africa. How things change. I’m not saying your wealthy democracy will become a totalitarian hellscape overnight, but I saying that whole nations do fall into sudden and shocking decline. The risk of utter catastrophe may be small, but it is real. So do everyone a favor: follow Masha Gessen’s . That said, Godwin’s Law probably still applies. Most demagogues are not autocrats, and most autocrats are vacillating narcissists. But even the least effective attract power-hungry hatemongers as lieutenants, and try to whip up frenzies against “enemy groups” as scapegoats and distractions. Don’t think you’re part of an enemy group, because you’re not a visible minority? Don’t be so sure! Technology is a fundamental power center these days, and autocrats . https://twitter.com/balajis/status/703503245390446593 This is precisely the rhetoric the Kremlin and its media lackeys uses against human rights organisations in Russia. — Michael Harris (@mjrharris) Whether you’re a tech titan who fears the next , or a coder worried the law (and/or your nation’s surveillance-industrial complex) may be selectively enforced against you — https://twitter.com/munin/status/798395821435867136 — be cautious. Use for messaging, and in place of email, where possible. Install . Encrypt your laptop and your phone. Turn on two-factor authentication for your email and other accounts. For more details, read: Did your opponent hack your nation’s electoral system to eke out a narrow win? If so, that’s too bad, but the great virtue of democracy is that it allows people to correct their mistakes. The single greatest danger you face is that the electoral process itself may become compromised, through voter suppression, good old-fashioned ballot-box stuffing, or compromised voting / tabulation machines. i don’t understand why we don’t verify machine tabulations with handcounts in random samples of polling places. — Steve Randy Waldman (@interfluidity) So: fight hard voting systems that don’t , and consider fighting reliable . Most of all, though, don’t think of the mechanism of democracy as a magical black box. Be ready to audit, inspect, and defend every part of it, if you think anyone might be trying to corrupt it. This is the most important thing you can do. That said, even if you avoid mass oppression and strife, even the “best” autocrats trigger countless low-grade petty cruelties that ruin double handfuls of lives every day. Worse yet, they also hamstring or corrupt the legal system, and many strip-mine whatever government safety nets may exist in favor of further enriching their kleptocratic cronies. In America, for instance, it seems reasonable to expect life to get measurably worse over the next four years for visible minorities, LGBTQ people, women who seek autonomy over their own bodies, etc. There are many for these people: Concrete steps for vulnerable individuals to consider before the Jan 17 inauguration: — Downtown Brown (@firecrackrbrown) …but maybe the most important thing is to include them in your communities, both online and off. Social media gets a lot of flak around elections, much of it justified, but it is also a crucially important substrate that people can use to band together and support each other when times get tough. This is no time to get all , or to write off anyone and everyone who disagrees with you as a monster. Note that the same people who say “everyone who voted for the other side is racist and cannot ever be associated with under any circumstances” also often say “everyone’s racist, it’s just a matter of degree, it’s implicit in the system in which we live.” Be very careful who you call an enemy. More us-and-them polarization is exactly what the autocrat wants. https://twitter.com/balajis/status/797528444884004864 More generally, if you want to win political battles, you need to erect broad tents. Do you hate libertarians? Get rid of that attitude, — and vice versa: 3.Libertarians should be directing energy at limiting power right now. Others who fear Trump should be allies, not targets of derision. — Radley Balko (@radleybalko) Some argue that we should be so inclusive that we replace representative democracy with , via movements whose representatives pledge to always be guided by their members’ online votes. In Italy, the very successful does just this; and, of course, people have proposed blockchain variants… I’ve been arguing for that the whole concept of a world partitioned into nation-states makes less and less sense. That may sound crazy … but I bet a lot of things are happening around you that would have sounded crazy before the election. I’m not saying you have to support or — though I also don’t think either should be automatically dismissed — but I saying you should do anything and everything you can, personally and technically, to strengthen and increase the density of supra-national networks. That means making a point of extending community and governance projects across borders. That means getting into Bitcoin and the other major cryptocurrencies, as the only true world currencies. That means doing your best to defend Internet (and other) freedoms around the world, where they are , by building decentralized systems that exist orthogonally to nation-states. It also means pilot projects wherein people begin to consider themselves as members, or even citizens, of formally constituted entities that span national borders. I know this sounds a little crazy. We are so accustomed to swimming in the water of a nation-state world that anything else seems profoundly alien. But it’s past time to at least tentatively experiment with new ideas. https://twitter.com/balajis/status/746543681079914496 At the same time, do everything you can to strengthen your own local community, in the streets and cities around you, and try to win as many political victories as possible at the city / country / grassroots level. (And if you’re bold enough to actually start a new political movement, for Zardoz’s sake start by trying to win small, and then work your way up, rather than the reverse.) Unfortunately the tech industry has been more focused on delivery apps for privileged hipsters than systems and networks that strengthen communities and create new tribes. Sorry about that. But it’s not too late to break some new and important ground. https://twitter.com/balajis/status/797525534150275072 There are interesting notions out there already. , for instance, a kind of fully voluntary micro-basic-income for small groups who vow to support each other in times of need through thick and thin. Based on blockchains, of course! And/or apps (although, personally, I think time banks would be far more successful; the precept that everyone’s time is worth exactly the same has historically proven unpopular.) Debug Politics is to work on political problems. (I know, I know, hackathons, ugh, but at least their intent is noble.) More generally, the tighter and closer everyone’s support network is, the better we will all be able to ward off the slings and arrows of autocracy. Think of it as the phalanx principle. Many, many countries have voted in autocratic clowns, monsters, or narcissists. It’s a mistake, but the beauty of democracies is that mistakes can, eventually, be fixed. Use technology to practice surveillance self-defense; proactively defend the mechanisms of democracy; build communities, both locally and across nations; and help the most vulnerable — and you’ll both limit the damage done and mitigate the gravest risks. As long as we stay stubborn, this too shall pass, and the arc of history bends towards justice.
|
Become Michael Knight with Dashbot, an AI for your car
|
John Biggs
| 2,016
| 11
| 26
|
Who needs a smart car when you have a cool car? The is a $49 add on for your vehicle that allows you to interact with your phone and Alexa while driving, thereby keeping your hands free to fend off pomaded villains and to swoop in an save innocent people before a helicopter explodes overhead. The product is a sort of Bluetooth speaker on steroids with lots of functionality built-in and the ability to swap from an Alexa-only device to a quick and easy way to interact with your phone without touching it. The system connects to your var via a cigarette lighter power port and the Aux in jack of your stereo or via Bluetooth. It connects to the Internet every time you bring your phone close to the car and because it is a self-contained computer you can even hook up a wireless OBD-II sensor and read car sensor data on the fly. The Dashbot uses the , essentially a miniaturized single-board computer like the Raspberry Pi. The chip runs Linux and is completely open source for your hacking pleasure. From the team: Because Dashbot is based on C.H.I.P. Pro, you can teach it new tricks. Dashbot runs Gadget OS, our mainline Linux based fast-booting operating system. Like all Next Thing Co. products, Dashbot is Open Hardware, its software is Open Source, and of course, Dashbot has an open API. In fact the Dashbot team is the same team that released the C.H.I.P. Pro and they also completed the to bring this product to life. They’ve far surpassed their goal with 20 days left in the Kickstarter and it looks like this could be a fun way to make an older vehicle just a touch smarter. [youtube=https://www.youtube.com/watch?v=wdKL_R0RNUA]
|
null |
Megan Rose Dickey
| 2,016
| 11
| 10
| null |
IBM expands its UK presence with 4 new data centers
|
Frederic Lardinois
| 2,016
| 11
| 21
|
today announced that it is launching four new data centers in the U.K. This brings IBM’s total data center footprint in the U.K. to six, in addition to 16 other locations across Europe. This is yet another example of the company’s increasing infrastructure investment. The first of these new U.K. locations in Fareham will go online in December, with the other three U.K. locations launching next year. As an IBM spokesperson told me, IBM isn’t building its own data centers, though. It’ll lease raw space and build out its own infrastructure (electrical, networking, compute, storage, etc.) in these leased spaces. The last new data center IBM opened in Europe was its Oslo location, in September, but the company also recently expanded its presence in Asia with the launch of its Korean data center in August. In total, IBM’s cloud services are now available in 50 data centers. IBM is betting on a number of different brands for these data centers. Its Watson cognitive services, for example, are only one of hundreds of different services that run in its data centers. Under the SoftLayer brand, it also offers more basic cloud servers and storage solutions, as well as its platform as a service offering. But there is also the OpenStack-based private cloud as a service and a that run on those different platforms. Looking at IBM’s cloud portfolio can quickly become overwhelming, but it’s also driving a good bit of IBM’s revenue. In its third quarter, cloud revenue grew to $12.7 billion. In the U.K., its marquee customers include the likes of Wimbledon, Dixons Carphone, Shop Direct, National Express and, as the company announced today, tour operator Thomson. It’s worth noting that IBM’s second new data center in the U.K. will operate out of an Ark data center. is a join venture between the U.K. government and the public sector. This will likely increase IBM’s footprint in the public sector in the U.K., as well. “By adding four new Cloud data centres in the UK, IBM is giving local businesses an easy route to the cloud, helping them quickly innovate and respond to market demands,” said Robert LeBlanc, senior vice president of IBM Cloud, in a canned statement. “IBM is continuing to invest in high growth areas, offering clients higher-value cloud data services such as Watson and Blockchain running on our cloud infrastructure that delivers world-class scalability, performance and security.”
|
Crunch Report | Oracle Acquires Dyn “Thanksgiving Special”
|
Khaled "Tito" Hamze
| 2,016
| 11
| 21
|
Tito Hamze, John Mannes
Tito Hamze
Gregory Manalo
Gregory Manalo TechCrunch C/O Tito Hamze
410 Townsend street
Suite 100
San Francisco Ca. 94107
|
Universal plans to bring Wattpad stories to TV
|
Anthony Ha
| 2,016
| 11
| 21
|
Social publishing platform is making its biggest move into Hollywood yet thanks to a new deal with Universal Cable Productions. UCP is a TV production company owned by NBCUniversal — it’s produced shows including USA’s , SyFy’s and Bravo’s . As a result of the deal, the company plans to develop new shows based on stories published on Wattpad — stories it says it will identify, in part, by getting access to custom Wattpad data. “Wattpad data will help spot compelling stories with massive built-in audiences, and fans will influence the development process,” said Wattpad Studios head Aron Levitz in the partnership release. “There’s no place for hunches and guesses when it comes to content development these days. Data- and community-driven entertainment is the future, and Wattpad Studios is leading the way.” UCP isn’t announcing any specific shows yet. A Wattpad spokesperson told me the company is looking to develop around five projects a year — but of course there’s no guarantee that any of those projects will make it onto TV. Wattpad says it reaches 45 million people each month, while its community members have published more than 300 million stories. This isn’t actually the company’s first foray into TV. For one thing, the TV5 network in the Philippines already adapted stories for . More broadly, Wattpad has signaled an ambition to bring its stories to other media thanks to and the launch of earlier this year.
|
Ears, hands or eyes
|
Michael Eisenberg
| 2,016
| 11
| 26
|
Even before smartwatches, I did not wear a watch. I did not buy an Apple Watch or Pebble, despite the cool factor, even though I always upgrade to the latest iPhone as soon as it comes out and the lines subside. I don’t like the way watches feel on my wrist; more importantly, I don’t get real value from a watch. Unless you are in the Army like my son, where a G-Shock is the preferred timepiece, there are (and for a long time have been) clocks on the wall everywhere. And there are phones with the time and date in every pocket. Needless to say, the graph below about the decline in smartwatch sales is no surprise to me. I would also suggest that in order for an innovative new device to take off and become a platform, it actually needs to pioneer a new user interface that utilizes a different one of our limbs or senses. The PC really took off with the GUI and the mouse. If you remember computers like the Commodore 64 before the mouse and the GUI, you know just how few of us there are who actually owned one and used one. The mouse put my in use and the PC took off. The smartphone, which was the next platform, pioneered the touchscreen at scale. There is a big difference in experience and convenience between my finger swipe and a full-handed mouse. This enabled a new experience, which, when layered onto the always-on and always-on-me nature of the smartphone, created the next platform. Apple’s iPod and its click wheel was actually the first indicator that the finger was another limb and a radically different UI than the mouse. The touchscreen built on that innovation. In contradistinction, the watch has the same finger-touchscreen interface. It does not tickle a new human sense, nor use a different limb. Hence, it does not create enough innovation to drive new applications, uses and, ultimately, a platform. There is another dimension though, as has pointed out (before me) — each new level or era of technology also extends human geographic freedom. The smartphone has certainly done this, as we can now do everything on the go; the internet did it before through remote information and services; and, before that, the PC expanded geographic freedom, as well, untethering people from the need to finish their work at school or the office. It is through the combination of these two frameworks that I think about the next generation of technology and areas of computing and applications in which I am hoping to invest over the coming years. I would suggest that when you look at the two platforms competing to be the next platform for the information revolution, this framework is useful. Facebook bought Oculus, essentially betting that VR is the next platform for computing. VR certainly makes use of another human sense: sight. It has a different user interface, my eyeballs, and a headset (or even perhaps my brain interpreting what I am seeing). However, I think VR fails the second test right now (and for the foreseeable future). It does not expand human geographic freedom. It actually constrains it. It is a sitting experience and it only virtually expands my geographic freedom. Virtual freedom is an escape — it is not actual geographic freedom. I think this, however, points us to where the real opportunity is: My and, by extension, my mouth. I think Amazon and Apple are onto something with Alexa and the wireless AirPods, respectively. Apple, the creator of the smartphone touch interface revolution, has intuited that Bluetooth, sensors, a wireless chip and other can be packed into my . Amazon, who whiffed on the Fire, has decided to leapfrog a generation and move to voice. Interestingly, while Apple (I think) got there by moving the device from my to my , Amazon (I think) got there because shopping is now an always-on experience and they want me to use my mouth to shout at Alexa whenever I discover that I am missing a product. Hence, the Alexa stands as virtual while the human uses his mouth while shopping to fill his refrigerator, and Apple uses my human to also free my and my sound-emitting mouth is a byproduct. Both Apple and Amazon, from whichever direction they approach the innovation, are using another limb: my . And by using voice as a command or interface, in proximity or over long distances, we make yet another meaningful extension of human freedom. Particularly, when we consider the fact that it is also a free interface. By freeing my , we can enable innovation that uses voice, sound and untethered in ways we cannot imagine yet. Imagine what human ingenuity could bring forward if we took our teenagers’ off their phones? Interestingly, this same observation about voice is likely what is driving Google’s Pixel phone. Google is not after the phone . We have heard from sources that 25 percent of the responses on Inbox are now smart-replies, which is quite incredible. This is in line with Google making sure to own the software and network layer of the interface of the future. That, in their view, is clearly voice. I am very excited about voice and voice applications, as well as in- wireless computing. I think it has another benefit, as well. By unshackling us from the little screens in our , it raises Homo sapiens’ currently-craned heads back to level. Then, we may also start talking to each other — and not just Alexa.
|
Bizly lets you book on-demand meeting spaces in luxury hotels
|
Fitz Tepper
| 2,016
| 11
| 21
|
Everyone talks about rooms being an underutilized asset in the hospitality industry. But what about meeting space? Almost all four- to five-star hotels, even boutique ones, have at least one meeting room. These can be 10-person conference rooms or 1,000-person banquet halls, but the concept is the same — it’s an empty space that the hotel wants business people to rent. But the process of doing so is inefficient — most hotels make you go through their events or catering department and fax forms back and forth until you agree on a price. So the result is that these rooms get used a lot less often than they should. is a startup trying to change this, by making it just as easy to book event space as it is to book a hotel room on a one-tap platform like Hotel Tonight. It’s pretty simple: Bizly will show you a list of hotels in your city with available event space along with an hourly price to rent out a room for a meeting — think a 20-person board room in the Ace Hotel for $75 an hour, or meeting space in the Mandarin Oriental for $150 an hour. Hotels are happy because they are generating revenue from a room that otherwise sits empty, and meeting organizers get to host a meeting in a luxury setting like a five-star hotel. And hotels still get to offer food and beverage to groups, which is traditionally where they generate revenue from meeting-room rentals. But by using Bizly to rent to small groups (instead of the traditional 200-person wedding or conference), hotels should end up increasing the utilization of their meeting space and maybe be able to reduce their dependence on food and beverage. So who would use a service like this? Bizly says that they’ve seen a ton of uses cases — venture capitalists needing a nice meeting space, startups that have overgrown their offices and need to hold an all-hands meeting and even corporate clients who need a reliable place to hold meetings when out of town. Plus, the startup prides itself on its last-minute nature — they say that most bookings happen less than six days in advance. So it really could be meant for a startup that needs to hold a last-minute meeting and wants to impress their guests by having them come to the Mandarin — all for about $150. Bizly launched an in February and in August, and has been tripling their bookings each month since. They have signed up more than 100 hotel brands in six cities, including pilots with Starwood and Hyatt properties. The startup also recently closed a seed round led by .
|
DeepMind Health inks new deal with UK’s NHS to deploy Streams app in early 2017
|
Natasha Lomas
| 2,016
| 11
| 21
|
DeepMind Health, the division of the Google-owned AI company that’s focused on building links to healthcare providers to drive the application of machine learning algorithms for preventative medicine, has inked a fresh data-sharing agreement with the NHS Royal Free Hospital Trust in London. The new data-sharing arrangement extends until at least 2021. It’s the second agreement signed between the pair — and it supersedes their original agreement inked last year, which ran into controversy after a freedom of information request by revealed the volume of patient identifiable medical data (PID) flowing from the Royal Free to DeepMind, and about whether NHS information governance principles were being correctly followed. The data in question was being used to power an app called Streams, built by DeepMind but using an NHS algorithm to generate alerts on patients at risk of Acute Kidney Injury (AKI). At the time the collaboration was made public, last February, no details were provided about how much PID was being shared between DeepMind and the NHS — leading to huge consternation when the scope of the arrangement emerged. The U.K.’s data watchdog, the ICO, began investigating complaints about the data-sharing agreement. The Streams app also ran into trouble when it was revealed DeepMind and the Royal Free with the oversight body, the MHRA, despite piloting the app in the Royal Free’s hospitals. The MHRA had not been approached prior to starting tests of the app. The pair subsequently suspended use of Streams in the hospitals. But they’re now announcing plans to restart the project — and, evidently, to try to reset it onto a firmer information governance footing. Above all this is an attempt to improve the tarnished public image of DeepMind’s inaugural push into preventative healthcare by trying to secure patient trust — to, ultimately, grease the future funnel for more data flows from the NHS to DeepMind. The point is, healthcare-related AI needs very high-quality data sets to nurture the kind of smarts DeepMind is hoping to be able to build. And the publicly funded NHS has both a wealth of such data and a pressing need to reduce costs — incentivizing it to accept the offer of “free” development work and wide-ranging partnerships with DeepMind (which has several ). DeepMind and the Royal Free confirmed today that the Streams app has now been registered as a medical device with the MHRA, and said it is ready to be deployed in the Royal Free’s hospitals from early next year. “Following prototype testing, as well as registration with the Medicines and Healthcare products Regulatory Agency (MHRA), this first version of Streams is ready to be deployed to clinicians across the Royal Free hospital sites early in 2017. It is expected to result in an immediate improvement in AKI-related patient safety and outcomes,” they write in a press release about what they describe as the “next phase” of their collaboration. There also looks to be a broadening of the scope, with the PR talking about expanding the app’s remit to cover early detection of sepsis and organ failure, as well as AKI. “The ultimate version of the Streams app will alert doctors and nurses to patients who need their attention in seconds rather than hours, reducing the number of patients who deteriorate in hospital without a clinician being aware,” they write, adding: “Streams will be extended beyond AKI to help care for patients with other serious conditions including sepsis and organ failure. At least ten thousand people a year die in UK hospitals through entirely preventable causes, and some 40% of patients could avoid being admitted to intensive care, if the right clinician was able to take the right action sooner.” On the information governance front, among the noteworthy developments are: Commenting in a statement, DeepMind co-founder Mustafa Suleyman said: “Privacy and trust are paramount, and we’re holding ourselves to an unprecedented level of oversight by publishing our agreements publicly and engaging nine respected public figures to scrutinise our work in the public interest.” Despite what is clearly a lot of re-engineering of the presentation and some changes in the structure of DeepMind’s collaboration with a publicly funded and much-beloved National Health Service, many questions remain unanswered — not least the core criticism that the volume of PID being shared without patient consent is questionable, given the pair have always relied on claiming they do not need to obtain patient consent for sharing the data because they say it is being used for what’s termed “direct patient care.” However, direct patient care refers to a direct care relationship between an individual patient and their clinician(s) — whereas some of the patients’ whose data is being shared under the Streams arrangement, so at least initially for the purposes of detecting AKI, will never be in the relevant direct care relationship because they will never develop AKI. Safe to say, the push toward “preventative” healthcare looks to be putting a lot of pressure on the NHS’ traditional information government processes — which are not set up for an era of big-data mining and machine learning-driven “future potential” promises. It remains to be seen whether the U.K.’s National Data Guardian will seek to provide some guidance here (following controversy generated by the original DeepMind/Royal Free data-sharing data, Caldicott has been between the pair). But as private sector giants like DeepMind make early bids for valuable public health data sets — for the etc. — governments and regulators have an equally pressing need to get their heads around the new reality of health data — as both a highly sensitive and personal resource and a commercial-accelerant-in-waiting that could enable the creation of a new generation of digital healthcare products. One thing is certain: Gaining and sustaining patient trust in any such systems will be essential.
|
UPDATED: Machine learning can fix Twitter, Facebook, and maybe even America
|
Chris Nicholson
| 2,016
| 11
| 26
|
Quitting Twitter is easy — I’ve done it a hundred times. Someone called it “a clown car that drove into a gold mine,” and like all clown cars, Twitter makes the passengers get out once in awhile. If I go back, it’s because I’m addicted. The tight news cycle, tweetstorms, gossip mongers, insight, argument, factoids, snark and one-liners. For an information junkie, that little bubble is hard to resist. But Twitter — and Facebook, for that matter — is desperately broken in ways that alienate users, spread hate and endanger us as a species. The elections have revealed how broken they are better than anything else could have. First, let’s talk about what’s broken. One set of problems are the collisions between unlike users, and the offense, outrage, and remorse that follow. Another, much larger set of problems arise from the falsehood, hate and lies that go viral on social media, and their electoral consequences. These two sets of problems are interrelated. We’re getting too much trolling and not enough facts: we need to screen out one and let in the other. The right filters can address both problems. People who have left Twitter in the last year, at least temporarily, include of the Ghostbusters remake, the British comic , and of A16Z. Other notable recent departures include , who was attacked after the suicide of her father Robin. That’s right: She was attacked on Twitter after the suicide of her father. They sent her fake mortuary photos of him. That’s an example of the first problem. A lot of people, especially in San Francisco, think that open communities are great and that social media should be all about connecting people. But not everybody should be connected. that television gave us the village idiot so that we could feel superior, while the Internet gave us the village idiot as a source of truth. Nobody wants to argue with the village idiot, let alone millions of them. On Twitter, you have to block them one at a time. That’s a lot of work, and by then it’s too late. Their trolling idiocy has infected your life. Their work is done before the rules can be enforced. Perfectly open communities always go sour. You need filters. Every functional community has them. And that’s where machine learning comes in. The natural-language processing to detect trolls, racism and insults isn’t hard, and Tweets as a data genre have been . We can build filters that work. (If you want to know how, you should read about .) Deep learning is setting new records in accuracy for a lot of difficult problems, including image and voice recognition. It will achieve similar gains with text classification using algorithms like and Doc2vec. If you can detect trolls, you can protect the people they’re trolling by muting or putting a warning over the trolls’ posts. Twitter could even figure out who likes a few threats of violence now and then and personalize the masking. Personally, I go on Twitter to learn new things and hear new voices. But there have been some interesting studies from that certain kinds of diversity can hurt civic life and erode trust. At the very least, it’s something that online communities should pay attention to, if they want to keep people coming back. There’s a radical openness to Twitter, which is cool some of the time, and uncool other times. It’s the uncool times that stick with you. You can’t unsee morgue shots of your father, like those that were Tweeted at Zelda. Twitter can do something about it, and they should. They already have a way of . Why don’t they do the same thing with ethnic slurs, death threats and other kinds of trolling? Just draw a curtain over them. After a while, people will figure out that they don’t really want to see what someone said, if Twitter masks it. And their day will be better. And they will keep using Twitter. “A lie gets halfway around the world before the truth has a chance to get its pants on.” Winston Churchill said that. With social media, a lie probably circles the world a couple times… The algorithms of platforms such as Facebook and Twitter may not shield us from hate, but they do encourage the spread of emotion-rich content among like-minded people, especially when that content triggers outrage. The more shares a post gets, the more it will be promoted to similar individuals because Facebook et al optimizes for engagement, period. Unfortunately, a lot of that content is false, and its popularity has consequences. One of the main problems with U.S. politics is a . It’s the replacement of a , as formulated in the years of George W. Bush, with a platform of wishful thinking … backed by nukes. That’s problematic for a lot of different reasons, notably the way it breaks our ability to understand cause and effect, trade, war and indifferent nature. It’s particularly harmful to how we relate to each other. Because facts are something that can unite very disparate people, while beliefs are endlessly divergent. Without them; bubbles all the way down. We live in an age of self-reinforcing beliefs, and the reinforcement of groupthink happens in a feedback loop with the media, especially social media. We need to stop the flow of hate and lies and help the spread of facts, because words matter. What do I mean by lies? I mean . And the weird way pumped out disinformation about Trump during this election cycle. Not only are our social media channels filled with garbage, but Americans are being gamed by foreigners. It should be illegal, but even if it’s not against the law, it’s something tech companies could control, if they wanted to… They can control it because we now have the ability to detect hidden patterns in text to — say — identify a book’s true author. (The pseudonymous author was outed by a statistical textual analysis of her work before an investigative journalist doxed her this year.) Just like Google can build a highly accurate spam filter to keep you from wasting time on the pleas of Nigerian princes, deep learning can classify text by many measures, including its degree of factuality, falsehood or truthiness. Algorithms can do that because we know how to “vectorize” text. That is, we can turn any text into a column of numbers, and those are called neural embeddings. It’s a simple, yet unlikely, translation to represent language in numbers. Doing that makes natural language computer-readable. Then we can perform powerful mathematical operations on text to detect patterns and similarities, make predictions and apply categories to it. Those categories might be: “probably false” or “probably true.” And once we know the likelihood of a text’s factualness, we can decide how far it should spread. We have fact-checkers at organizations like , or applying judgments to news stories already. Those could be turned into labeled datasets to train algorithms to categorize text they’ve never seen before. If that’s not neutral enough, Facebook could build its own team of fact-checkers. The real question is, do the tech companies want to control it? Mark Zuckerberg is still thinking about that one. Facebook and Twitter flattered themselves that they played a , but Zuckerberg said this weekend that it’s a “ ” that fake news on Facebook affected this tight election. You can’t have it both ways. A smug, amoral response from the people at the top of powerful tech companies isn’t what we need. They have a responsibility to the public, to the species and to themselves to promote the facts and to mute the hate and lies, even if that responsibility is not enshrined in law. Not least because Mark Zuckerberg is Jewish, and Donald Trump rode a of and to power. It doesn’t matter how you identify when they start handing out the yellow stars. While meant diddly squat this election cycle, the way that media and social media promoted false stories week after week to increase their eyeballs and mindshare had a huge effect. This presidential election tipped on a couple percentage points in a few key states. Or to be precise, in Wisconsin, Michigan and Pennsylvania handed America to Trump. That’s the equivalent of the population in Boulder, Colorado; West Palm Beach, Florida; or Daly City, California. Do you think , the months of , or Wisconsin’s might have accounted for a town’s worth of ballots? If so, why wouldn’t the algorithms of a powerful social media platform used by tens of millions of US voters? Broadcast media spent covering the non-scandal of Hillary’s emails as it did covering the issues, or her policies, and social media amplified instead of remedied that distortion. Some great journalists wrote some great stories this year. They reported on Trump’s , his virtually , and his to . But those stories didn’t have legs. They never reached the larger audience that needed to hear them most, because we have become polarized. We go looking for opinions that agree with ours. Each of us needs some windows opened onto the disagreeable facts and inconvenient truths that will slap us in the face no matter what we wish for. The tech platforms powering social media can help reconcile us with reality in many quiet ways, or they can join the that ushered a , a bigot and a into the White House for the sake of an earnings report. Until then, what we read on Twitter and Facebook will add nothing to our understanding of the world. It will just be our own breath backing up on us. And on that note, I think I need a Tic Tac.
|
The NTSB is investigating the ‘structural failure’ of Facebook’s Aquila internet drone
|
Devin Coldewey
| 2,016
| 11
| 21
|
Say what you will about the merits of Facebook’s Internet.org and Free Basics — it’s pretty cool that they’re building a to deliver it. But a “structural failure” that occurred on the Aquila’s first test flight may be more serious than Facebook made it out to be: . The NTSB confirmed this and provided further details. Facebook (which occurred on June 28) in July, listing several things they were looking at, learned and so on. Under the “Real-world conditions” bullet point, admits things weren’t nominal: We are still analyzing the results of the extended test, including a structural failure we experienced just before landing. We hope to share more details on this and other structural tests in the future. They didn’t, possibly because of the NTSB investigation, but Facebook did issue a statement today emphasizing the positive outcomes of the test: We were happy with the successful first test flight and were able to verify several performance models and components including aerodynamics, batteries, control systems and crew training, with no major unexpected results. Really, it was too much to hope that nothing would go wrong on the first full-scale test of an enormous, experimental aircraft design. A source close to the project told TechCrunch that some damage was expected, since the Aquila isn’t actually designed for repeat takeoffs and landings (it has skids, not landing gear), and also because the day was windier than expected. The failure occurred just a few seconds before landing from the craft’s 90-minute flight, the source said. It’s the NTSB’s prerogative to investigate any airborne troubles like this, and clearly it decided to so in this case, perhaps because of the high-profile nature of the test and aircraft. But the NTSB wouldn’t get involved if a screw dropped off: a representative explained that it investigates when aircraft weighing 300 pounds or more cause death or serious injury, or incur “substantial damage” — defined as damage that “compromises the airworthiness of the aircraft.” That said, if the Aquila had nose-dived into the ground, caught fire or sustained some other high-profile damage, that likely would have come out by now. A full report is expected in a month or two, at which point we’ll have more details — but considering the scope of the project and pride evinced by Facebook in the Aquila’s development, it seemed reasonable to, well, clip its wings a little bit.
|
Po 3D prints personalized prosthetic hands for the needy in South America
|
Devin Coldewey
| 2,016
| 11
| 21
|
is advancing in two directions: on one hand, if you will, the latest in soft robotics and shrinking sensors enables limbs and digits with ever-increasing levels of sensation and realism. On the other, rapid manufacturing techniques are making it possible to bring sophisticated designs to the underprivileged and geographically isolated. pursuing the latter goal, making customized prostheses for people who might otherwise have never gotten anything at all. Paraguay, as described to TechCrunch by Po co-founder Eric Dijkhuis, is “a country full of amazing people but also a lot of challenges. There is a high number of amputations per day, upper limbs being a high percentage of them, due to a lack of work safety regulations combined with vulnerable working areas, and a lot of motorcycle accidents.” Widespread low income also means very few people can afford the kind of prosthesis they need — less than 3 percent, Dijkhuis said. Why should that be, Po’s founders thought, in an age when you can print objects into existence and get advanced control systems off the shelf? https://www.youtube.com/watch?v=7XoglCG6fLE So they got to work designing a durable, printable prosthetic hand and lower arm that can be tweaked and customized in terms of size, color and other basic parameters. There are over 100 Po hands out there being used right now that are controlled mechanically, but the team had a eureka moment when they met with about that company’s Myo device. : It fits around your arm, monitoring your muscles for bioelectric signals that coincide with various motions and passing that data on wirelessly to other devices. So making a fist or tilting your hand up could, for example, minimize the window on your laptop, or switch applications — or, in the case of a prosthesis, simply mirror that action with the artificial limb. Now Po has five people testing MyPo, a combination of the original mechanical arm and a Myo control mechanism. “At a fraction of the cost, MyPo mirrors the traditional functionality of a prosthetic hand,” said Dijkhuis, “including several grips, degrees of freedom, and it can even integrate applications that already exist with the Myo armband.” So in addition to picking up and manipulating objects, gestures can be built into interact with social media, music apps and so on. This last is a capability being investigated by others, as well: a Myo-powered accessory that fits onto existing prostheses and allows for this kind of interaction. An advantage of using the Myo as a limb control mechanism is that it can tie the motion of the hand directly to the gestures and muscular activity already learned for that motion. So although the user may not have fingers to make a fist, their motor memory of the way they did it remains, and the Myo can detect it and respond. “They think, they remember, they do, and the Myo armband guides the whole process amazingly,” said Dijkhuis. Po isn’t the first or only 3D-printed prosthesis, of course — many have been made already. But it’s not enough to simply make a design. There’s the matter of fitting, configuration and the cost of the components. “We we help the user pay a figurative amount that they can afford by subsidizing the rest through private donations,” Dijkhuis explained. “We also work with independent professionals, NGO’s, allied businesses and public organizations that support our work. Our model is currently being replicated in North Argentina and South Brazil by Po partners, but our entire workflow is going to be open so anyone can start their own chapter with standardized, ready to go procedures.” Meanwhile, , where you can download it, tweak it, make suggestions or perhaps try one out for yourself. This isn’t a particularly high-profile application of technology, since the users aren’t tech-savvy urbanites and the companies and services don’t command billion-dollar valuations. But getting a prosthetic hand to a child too poor to afford one is a goal worth shouting about. “Creating and developing Po, we have seen the power of leveraging new technologies, like 3D printing, the Myo armband, and the power of open source,” said Dijkhuis. “We believe that these technologies applied to social impact are not only disrupting an industry, but are rewriting the rules of the game for the future of prosthetics, and handing the power of innovation to people all around the world.”
|
This 18-year-old just raised $3.5 million to help developers easily add capabilities to their apps
|
Connie Loizos
| 2,016
| 11
| 21
|
, a two-year-old, San Francisco-based startup that enables developers to find, test, and manage many of the APIs they want to integrate into their apps, has raised $3.5 million in seed funding. The round was led by Martin Casado at Andreessen Horowitz, with participation from FundersClub, SV Angel, Green Bay Advisors, and 500 Startups, whose accelerator program it recently passed through. The company is interesting for numerous reasons — not least of them its founder, Iddo Gino, an 18-year-old who was a high school senior in Haifa, Israel last year when he was “discovered,” so to speak. It happened at a local hackathon, where Gino caught the attention of Dov Moran. The renowned Israeli angel investor provided Gino with some early mentorship, as well as $250,000 in seed money that Gino — who studied briefly at the public research institute – used to move to the U.S. in January. Of course, RapidAPI’s mission — to enable developers to access and easily use publicly available APIs in a way that allows them to seamlessly communicate — is obviously of great interest to investors, too. Gino likens these functional “blocks” that can be woven together, to pieces of plumbing, explaining that, “Each company has its own language. So [a developer] tries to learn how Facebook speaks then uses a dictionary to translate it back to English. And [he or she] has to do this every time” with every company and sometimes even across one company’s different divisions. RapidAPI has meanwhile created what are essentially big repositories of translator units, “so we don’t need an API company to change anything. We take care of turning everything into a single language.” Indeed, among the trends on which RapidAPI is capitalizing are so-called micro services, where instead of creating hard-to-maintain software, companies are take their existing software and infrastructure and turning it into APIs. Amazon has famously been focused on this issue for the last , expressly so its different teams could communicate with each other through these interfaces. Then again, Jeff Bezos tends to be ahead of the pack. At most companies, APIs have sprung up in a more willy-nilly fashion, where one team within a company might be working on a chat service, another on image uploads and cropping, and a third on service management, but none of the APIs they’ve developed are particularly adept at working together. At eBay, for example, hundred of APIs had been created over time with different standards and protocols – all of which were speaking very difficult languages that made it difficult for them to talk with one another. RapidAPI — which offers free and open-source adapters to connect APIs but plans to make money as a kind of services integrator — is now helping the company manage and connect those many internal APIs. And RapidAPI has something else going for it, too — an apparent opening in the market. For example, some developers have likened the technology to Parse, a toolkit and support system for mobile developers that Facebook acquired in 2013 and earlier this year when it decided to focus its attention elsewhere. Meanwhile, other beloved API management platforms have also disappeared into big acquirers in recent years, including Apigee, bought in September by Google for , and Mashery, acquired by Intel before being sold to Tibco. RapidAPI says it supports more than 200 APIs, currently. Gino says the 16-person company “wants to multiply that by 10” in a year’s time. In the meantime, among the most-used APIs by developers who visit its platform? The Spotify API and the Google Search API.
|
Why now, more than ever, we need a Twitter that works
|
Rishi Garg
| 2,016
| 11
| 21
|
, to paraphrase John Oliver, . Much of it took place on social media, and much is being written of late about how Facebook and Twitter have changed virtually everything that we’ve come to see as a normal part of an election cycle: the consumption of daily news and facts (or non-facts). Remember 2012, when due to Obama’s skill in exploiting the fledgling platform? It seems almost quaint now. 2016 will be remembered as the election where the power of Twitter was revealed in full as a platform that can enable an individual with little formal organization to bypass media institutions and speak directly to a populace all the way to the presidency. On , Trump could hardly contain his self-satisfaction regarding how well Twitter and Facebook had served him. Why was exploiting Twitter such an advantage for Trump? Because Twitter is an unusually powerful media platform, providing an unprecedented ability to reach anyone in the world, with far less friction than ever before. That ability points to a deep responsibility for Twitter and Facebook, among others, to acknowledge their roles as arbiters of information that can have historical implications. And it’s possible Twitter and Facebook aren’t up to the task — there is a need for a responsible social network, and if they can’t do it, someone else has to. Consider how much has changed in nine short years. Until 2007, if you wanted to build an audience and sell your ideas to millions of viewers, you needed to raise money to build a cable network or a printing office, hire a sales force to sell ads or hire a bunch of Columbia journalism or USC film grads to produce content. You needed to buy cameras and pay business affairs people and all manner of other things. Not anymore. Twitter has given anyone and everyone a direct voice to the world by smushing together three things that in traditional media have been mostly separate: or a platform to connect the world (formerly Comcast or Dish; today the completely commoditized mobile phone and internet); (formerly the 30-minute video-on-a-screen-plus-ads model; today the pixels that comprise an app like Twitter or Facebook); and (formerly Rachel Maddow’s show; today tweets and Facebook posts). The implications of this are profound. First, Twitter owns its own distribution. Because it’s a network, every new user makes the platform that much more essential, whether you’re one of the Monthly Active Users that Twitter is so often maligned for not adding fast enough — or whether you’re one of the hundreds of millions of people who are consuming news elsewhere about what’s happening on Twitter. Traditional media amplification of Twitter content doesn’t get Wall Street excited, but it nonetheless solidifies Twitter’s place in the world. Prior purveyors of the pipes to consumers (like cable companies) had neither the leverage nor the capabilities to play the role of content arbiter with subscribers. Next, because the application is owned by Twitter, the company has as much power in determining how people consume and create that content as the earliest creators of movie cameras, network television and production houses had in the 1930s when TV was just becoming a thing. We take these now-“traditional” formats as gospel, but they were merely invented by normal people in charge of the early medium, and they persist to this day. Similarly, Twitter’s founders (and the communities they enable), by intentionally designing the user experience that we all consume, shape what gets out of our phones and into our brains. More than ever before, the medium shapes the message. And finally there’s the content: free, live, visceral, formatted specifically for this application and ready to deliver 140 characters, photos, videos, Vines (RIP), etc. of noteworthy and not-so-noteworthy stuff right to your phone, computer or (Apple) TV. So new platforms unprecedentedly combine Comcast, Sony and MTV into a single, powerful platform controlled by one team. Gatekeepers are removed at every step, and anyone can participate. That’s why owning a share of Twitter (or Facebook, Snapchat or Instagram for that matter) is like owning a share of an entire media ecosystem, indeed like owning the whole , not merely a single company or player within it. Historically, with every other new media platform since Gutenberg invented movable type, once a new media platform is created (books, newspapers, magazines, radios and television), the first major media starts broad, attempting to bring as many people as possible to the new platform (like the original “Big 3” broadcast networks). Then, over time, once people are used to the new platform, verticalization occurs to super-serve certain demographics or constituents, like Nickelodeon serving kids or MTV serving teens. People self-select into the various cable channels, but the major broadcast networks remain to serve a mainstream audience a more diverse set of perspectives. But with social media you have an added dynamic: personalization. No one person’s Facebook or Twitter feed looks like anyone else’s. There’s little room for an MTV to come and super-serve the teen demo on Twitter: Teens usually do it themselves by creating a follower graph that serves their needs. If they can’t find what they’re looking for, they switch applications entirely (hence, in part, the rise of Snapchat). This means that whether users actively want access to a very limited stream of information, or if they simply engage more with a certain kind of content, they end up with a far narrower information stream. My friends from my poor hometown in South Texas see a very different Facebook News Feed than I do, even without active curation, because they engage with only a certain kind of story and, in turn, Facebook reinforces that interest. While Twitter’s real-time stream invites some amount of surprise and diversity into the experience, it’s nothing like knowing that we all saw the same nightly CBS/NBC/ABC news show, regardless of geography. So Twitter and Facebook today are wholly owned news and information platforms, praying mainly at the altar of increased engagement, with personalized, increasingly limited information streams, no embedded gatekeepers and completely open participation. The 2016 election and Trump’s victory in part show how powerful this democratization can be, because when one masters Twitter, one can impact public discourse as much or more than can the highest-rated TV network. For example, Fox News has a 1-2 million total viewer count on its best day. Trump’s Twitter account reaches 15.3 million people every time he says something. This kind of reach makes a platform like Twitter very hard to unseat. But when we rely on the community of users and followers to be the gatekeepers, to take the place of all those trained and experienced folks who have made up the media institutions that we’ve relied on for so long, we can also see what’s missing — and the consequences of omission are severe. As many have noted, conversation about the election was deeply marred by rapid sharing of false truths and misinformation, with no filter. Because most people today get their news from social networks, this is deeply troubling. The platforms, Twitter and Facebook among them, have to take responsibility — because claiming neutrality at this stage, hiding behind the “technology company” label, when they represent such platform-level power, is absurd. It comes down to two things: Could emerging entrepreneurs create a new social platform that combines the expression, creativity and easy trading of social currency that we see from Facebook and Twitter, but with an eye to more thoughtful discourse? Something that captures the clear need we have as citizens to develop and express identities around news and information, but with built-in means to edit bad information and non-constructive conversation? Even given the enormous network effects inherent in the major social media platforms, creating a new one isn’t quite as ludicrous as it may sound. If Facebook and Twitter are, indeed, new whole-cloth media platforms, then we’ve witnessed the creation of no less than such global platforms with more than 100 million users (and in some cases, a billion or more users) in the last decade alone: Facebook, Twitter, Instagram, Snapchat, WhatsApp, Telegram and Pinterest. And I’m not even including the massive Asian platforms, such as WeChat and LINE. When big challenges present themselves, we look to both the established players and emerging entrepreneurs to take big risks to make things better. I’m excited to see how this moment in time galvanizes entrepreneurial attention on making social networks work in our new world.
|
Google acquires Qwiklabs to teach developers cloud skills
|
Frederic Lardinois
| 2,016
| 11
| 21
|
Google today that it has acquired , a hands-on learning platform for those who want to become more familiar with operating cloud environments and writing applications that run on them. Qwiklabs, which launched in 2012, has only on teaching skills for Amazon’s AWS platform so far. Given AWS’ dominance in the marketplace, that made perfect sense. Amazon even uses Qwiklabs as its for offering self-paced labs for developers on its platform. Google says it will use Qwiklabs’s platform to focus “on offering the most comprehensive, efficient, and fun way to train and onboard people across all our products on Google Cloud, including and .” Qwiklabs says it will continue to offer lab learning credits and subscriptions on its site and it looks like the existing AWS labs will continue to function. I’m not sure we’ll see all that many new AWS courses in the future, though, but that remains unclear. A Google spokesperson told us that the company doesn’t have any further specifics to announce about this. As far as Qwiklabs’s Google Cloud courses go, the same spokesperson also told us that the company doesn’t have any specifics to announce as to when those will go live, either. Qwiklabs says that over half a million users have used its platform to spend over five million hours of learning about AWS so far. , the company never raised any outside funding and neither Google nor Qwiklabs disclosed the price of today’s acquisition.
|
Overclocked smartwatch sensor uses vibrations to sense gestures, objects and locations
|
Devin Coldewey
| 2,016
| 11
| 21
|
Just from patterns of motion, your smart devices know when you’re walking, when you’re riding a bike and when you lift your wrist to check the time. But it turns out they can also tell when you snap or make a fist, or whether you’re holding a smartphone or steering wheel. All they have to do is listen a little harder — well, about 100 times harder, actually. Researchers from Carnegie Mellon University created that supercharges an ordinary smartwatch’s accelerometer, allowing it to sense incredibly tiny variations in vibration frequency. That could be the thrum of an engine, the note being tuned on a guitar or the slight differences apparent when you move your hand in different ways. The secret lies in the specs of the accelerometer itself. Normally, they sample motion somewhere around 20-100 times per second — more than enough to tell whether the user is walking or running, for instance. But CMU’s Chris Harrison and his colleagues noticed something. “Right there on the data sheet, it says ‘maximum speed 4,000 Hz,’ ” said Harrison in an interview with TechCrunch. It was capable of polling motion more than a hundred times faster than any smartphone was telling it to. “We saw that and said ‘hmm, bet there’s some interesting stuff there.’ ” Sure enough, there was. Even when propagated through “a water-filled sack of bones,” as Harrison described the body, just about everything produces a unique high-frequency vibration pattern, a sort of acoustic signature that can be used to identify it almost immediately. At first, the team considered use cases such as being able to snap your fingers to turn on a light. “But that seemed kind of gimmicky,” he said. “More interesting is being able to use the arm as an extension of this sensor. We can actually detect what object you’re grasping as soon as you grasp, and we can detect whether you’re in the car, in the kitchen…” Think of a timer that turns on as soon as you grab your toothbrush, maps that appear when you touch the map at the entrance of a building or a two-factor system that checks not just whether you have a device, but whether you’re sitting at your desk. It can be combined with active electrical and wireless signals sent out by the watch to strengthen the recognition process. “Various people in the industry have reached out to us; they’re like, ‘huh, didn’t know we could do this.’ We’re in talks with some people right now,” said Harrison. “The capabilities we show, Apple or Samsung or whoever got on board could deploy it — it’s all software.” Examples of signals produced by various objects when pinged by the smartwatch. “I mean, there’s a reason we hacked an Android watch,” he added. “All these watches have accelerometers and they all have high-speed modes. But we couldn’t do it on an Apple watch, it’s very hard to hack.” This isn’t the first foray into augmenting the capabilities of smartwatches made by Harrison’s lab. has shown the possibility of wirelessly detecting , or and hand the watch is attached to. “Smartwatches have to have like a one-inch screen,” Harrison said, “so how do you expand the envelope of interaction without sticking a giant screen on it? That’s research we’ve been working on for about five years.” The team’s work was selected to receive one of four “best paper” awards at the .
|
Genalyte raises $36 million from Khosla for its one-drop blood test
|
Sarah Buhr
| 2,016
| 11
| 21
|
Theranos competitor just announced it has raised an equity round of $36 million in funding led by Khosla Ventures and Redmile Group. The San Diego-based startup leverages lab-on-a-chip technology it says can run through up to 128 different tests on a single finger prick of blood in under 15 minutes using what it calls its Maverick Detection Platform. Genalyte’s proprietary silicon chip contains arrays of photonic microring sensors to detect diseases like rheumatoid arthritis and the startup hopes to use this technology in an outpatient setting, pending FDA approval. Genalyte has already gained permission from an Institutional Review Board (IRB) to run aon the accuracy of its finger prick technology against whole blood testing techniques and plans to use the new funding to run more clinical trials in the near future to gear up for possible regulatory approval. So far, CEO Cary Gunn says , which were presented at the American College of Rheumatology (ACR) annual meeting on November 12, have yielded positive results and the company is now ready to take it’s technology to the next phase. Genalyte’s technology is already used commercially in the pharmaceutical industry but, as mentioned, Gunn wants to move it to a “near-patient” setting where doctors can run diagnostics on a prick of blood right at their practice and get the results within minutes, as opposed to sending patients out to a lab and having them wait a few days for results. This is somewhat similar to Theranos’s original plan when it worked with . In theory, patients would go to a Theranos lab within Walgreens, give a finger prick of blood and get results on an app later that day. However, the industry is skeptical — especially in light of Theranos — and Gunn says he plans to avoid the many missteps of its competitor with a bevy of clinical trials to ensure the technology works before releasing the technology to consumers. The company has already published its findings in scientific journals — something Theranos did not do. It was also started by those with a medical background and involved the medical community from the beginning — another thing Theranos was criticized for not doing. “[The industry] wants to see data. They want to see studies done and they want to see results and that’s the way we are engaging with them and that’s something that takes time,” Gunn says. “It’s not always the sexiest thing for the print media.” While some the beleaguered Theranos was our only hope for a breakthrough in blood diagnostics (and that the press ruined any prospects of progress in this endeavor), Genalyte and others like it show promise in the field. We had an opportunity to give 2 drops of blood instead of a gallon and you shut it down before she had a chance! — Jesse Draper (@JesseDraper) “The diagnostics industry is on the cusp of a sea change and Genalyte is the standard bearer in an effort that will transform not only diagnostics but how precision medicine is practiced,” head of Khosla Ventures Vinod Khosla said of his investment. “At Khosla Ventures, we’re proud to expand our partnership with a firm that has and continues to demonstrate a commitment to a rigorous scientific process in order to provide timely and accurate access to diagnostic and health data.”
|
Munchery appoints new CEO amid reported struggles
|
Megan Rose Dickey
| 2,016
| 11
| 21
|
On-demand food delivery startup has a new chief executive officer. , the former CEO of Simply Hired, is taking the place of Munchery co-founder Tri Tran, who is now the company’s chief strategy officer, the company announced today. The appointment comes on the same day that Munchery has struggled to improve its margins, makes more food than it sells — and ends up either donating or throwing some of it away — and spends too much money on marketing. The on-demand food company wastes an average of 16 percent of the food it makes, according to internal documents obtained by Bloomberg. Munchery has also reportedly spent “hundreds of thousands of dollars” just placing on doors flyers offering discounts. But Tran said in a statement to TechCrunch that Munchery donates at the end of the day the food it doesn’t sell, and ultimately tries to keep overproduction of food to a minimum. Munchery doesn’t waste food. Anything not sold is donated at the end of each day to organizations that deliver them to the homeless and other residents in need. We take deep pride in this. We try to keep overproduction to a minimum. We have made progress as a startup but we are always trying to do better. The proper amount of production is a challenge that all food businesses deal with. Restaurants, catering facilities, and even online platforms like ours have to determine how much food to produce ahead of demand. We predict customer demand on a weekly basis, and execute our menu ahead of customer orders. We are continuing to try to strike the right balance. But tumult is nothing new for Munchery. In March, , La Boulange founder Pascal Rigo, just five months after joining the company. At the time, Rigo said he left because he didn’t share Munchery’s strategy for growth and vision for the future anymore. In a recent interview with Bloomberg, Rigo elaborated, saying that he didn’t think it was worth the time “to chase artificial growth alone.” He also said that Munchery lacked transparency around customer churn. In the last year, Munchery has tried a few different strategies — from to . In April, Munchery launched its second go at a membership program in order to try to increase its addressable market, Tran told me at the time. For what it’s worth, Munchery isn’t the only food company struggling. And, to be honest, struggling in some way or another in this space seems to be more of the norm than the exception these days. Blue Apron, for example, , according to a report by BuzzFeed. Meanwhile, Munchery competitor Sprig paused its delivery service in Chicago, let go its drivers and kitchen staff in Chicago and laid off seven members across its marketing and operations teams. Though, in Sprig’s case, . Munchery has raised about $120 million and is looking to raise more, according to Bloomberg. Munchery says it is profitable in San Francisco, and “contribution margin positive” in its three other markets. “One would be hard pressed to find a competitor that could make these claims,” Tran said in a statement to TechCrunch. “And even more importantly, our customer base and revenue are growing.”
|
Trump transition team appointments indicate a bid to dismantle net neutrality
|
Devin Coldewey
| 2,016
| 11
| 21
|
President-elect Trump has made it clear that he is towards the FCC’s implementation of net neutrality, both in his own words and, today, the of two long-time adversaries of the policy to his transition team. Jeffrey Eisenach is an economist and government veteran who worked at the FTC in the ’80s; he’s worked for a number of think tanks and research institutes that transmute industry money into custom expert critique, and under their auspices was a vocal opponent of the FCC’s current net neutrality rules. did some excellent reporting on the man back in August, if you’re curious about the possibility of conflict of interest. Eisenach described net neutrality as “an effort by one set of private interests to enrich itself by using the power of the state to obtain free services from another” in his . He suggested ISPs have no reason to discriminate between services, and they engender innovation rather than stifle it. You may judge that idea on its own merits. Mark Jamison worked on Sprint’s lobbying team in the ’90s, and like Eisenach has done expert consulting work for several organizations. In a recent op-ed, a “growing miscellany of ex ante regulations that frequently work against the entrepreneurs and consumers the rules are intended to help.” He sums up his argument, such as it is, here: He and Eisenach are both “scholars” at the American Enterprise Institute, a non-partisan think tank (I put that in quotes because while both are in fact actually professors, the term scholar in that context is a bit difficult to pin down). They also both contribute regularly to , a right-leaning tech policy blog composed mainly of grousing at the inefficacy of the current administration and organizations like the FCC. The FCC itself declined to comment on the appointments or whether they would be working with the Commission in any official way, but even if Eisenbach and Jamison are strictly advisory, they are in a position to exert quite a bit of leverage. How Eisenbach in particular can be considered not to be a lobbyist, however, by an administration ostensibly critical of employing them, is something of a mystery. With the current congressional makeup and a President-elect hostile to the idea of net neutrality, it’s probable that the rules enacted by the FCC have little time left. Republican legislators opposed the FCC rules but faced the prospect of an Obama veto if they attempted to pass a law nullifying them or stripping the FCC of its authority to regulate ISPs as it has. At this point it seems to be only a matter of time before the rules are rolled back.
|
Reelgood debuts a hub for finding TV shows and movies across streaming services
|
Sarah Perez
| 2,016
| 11
| 21
|
More people are cutting the cord with pay TV in favor of streaming services, but finding something to watch is growing more difficult – not only because of rights deals which see content coming and going, but also because there’s now a wider array of services to choose from. A startup called wants to help, and is today launching a centralized platform for web users, where they can find and watch content from multiple services from a single destination. The company an offering a social network focused on movies. However, the new web service is attacking the broader problem of content discovery across platforms, including both movies and TV, as well as offering tools to track the shows and films you want to see. also recently launched an Apple TV app which functions like the website, and is even a bit further ahead, in terms of feature set. At launch, the new site aggregates the content from a handful of top streaming services: Netflix, Hulu, HBO (HBO GO and HBO NOW), Amazon Prime Video, Showtime, Starz, and FXNow. But Reelgood founder David Sanderson tells us they have 28 different sources ready on the backend, and will listen to consumer feedback to see which ones to roll out next. The website itself is easy to use. Content is split up into “Movies” or “TV” sections, and there’s a search box if you’re looking for something more specific. Here, you can search by title, but Apple TV supports searches by actor and director as well, which is making its way to the web. Within each high-level category, the available movies or shows are broken down by genres, like comedy, drama, documentary, mystery, horror, and so on. But Reelgood’s web version is still lacking a little polish. For example, longer content descriptions are currently cut off – another thing that’s being addressed in a later release. On each item, you can see both the critics’ and audience’s rating from Rotten Tomatoes, along with other information, like the description, length, and rating. The Apple TV app shows more information here, like director and cast. On each item, you can mark things as “Seen It” or add them to your universal watchlist. This becomes tiresome to do individually – it would be useful to do this from the main page. This, too, is built, but wasn’t ready at the time of launch, says Sanderson. In addition, there’s a list of “similar” content suggestions underneath each item, but the recommendations seem to be a little off. For instance, how is the horror movie “Seven” related to a movie like “Paper Towns,” which is more of a coming-of-age story involving kids looking for their missing friend? The connection between “Grease” and “Straight Outta Compton” is that they’re movies about music. (Clearly, this needs some work). For users of the earlier iOS app, you’ll gain access to social features where you can see what friends are watching thanks to their marking things as “Seen,” but new web users will need to wait on this to arrive for them. In time, the iOS app will be revamped to mirror the web service, though Reelgood’s more immediate focus is on where people are watching: web and TV. Apps for Roku and Amazon Fire TV will likely hit ahead of new mobile releases. Though still a beta product, as a basic web aggregator, Reelgood works. When you find something you like, you can watch a trailer or simply click “play” to be taken directly to the streaming service’s page for that title to start viewing. Over time, the company plans see a path to revenue through affiliate incomes for subscription sales to the third-party streaming services, and, in the longer term, by aggregating and reselling the data on viewing behavior. “Nielsen is only in 150,000 homes…whereas we hope to be ubiquitous across all different platforms, and have way more data,” explains Sanderson. “We know exactly what people are watching and what they want to watch, what they like, and what they didn’t like,” he says. Reelgood’s team of seven includes the entire development team from the streaming app for torrents, Popcorn Time, along with founder Sanderson, previously a PM at Facebook. The startup earlier raised $1 million from Harrison Metal, and is now working to close a $1.5 million seed round from Harrison Metal, Social Capital, and the President of Warner Bros. The service is now, while its iOS and Apple TV apps are available in the respective app stores.
|
Ava gives the deaf and hard-of-hearing a more present voice in group conversations
|
Lucas Matney
| 2,016
| 11
| 21
|
For those with hearing issues, simple dinner table group conversations can be pretty painful to stay on top of. is aiming to bring deaf and hard-of-hearing people back into group conversations with their threaded speech-to-text application that gives people with hearing issues an easy way to stay on top of a conversation. The app, formerly known as , starts with each participant in a conversation downloading the app and setting up a profile. This may seem like a bit of work to get a casual chat going but especially for families or groups of friends, it’s a really simple way to bring everyone into a conversation regardless of their specific hearing abilities. After getting everyone onboard, people just talk normally near their phones microphone and the speech to text translation is organized into a threaded message for everyone in the group text allowing users who are deaf or hard-of-hearing to have a record of the conversation right in front of them to quickly respond to. The company this past week that they’ve closed $1.8M in funding from a host of investors including SV Angel, Lerer Hippeau Ventures, Partech Ventures, Steve Blank and Tim Draper. I sat down with Ava CEO Thibault Duchemin to chat about tech meeting the needs of the deaf community and what’s next for his app. The potential for the app expands well beyond dinner table conversations, and fits into pretty much anything where multiple people are chatting in an environment where deaf people are present. The Ava team had a major opportunity to show off their tech last month when Salesforce used the tech to live transcribe the majority of its breakout sessions for audience members at its Dreamforce conference. Duchemin and COO Pieter Doevendans also chatted a lot about the potential for education markets in giving everybody in the classroom an equal voice. For the most part, speech recognition tech has been slow to wide user adoption because it’s not perfect. Even when systems can handle 90 or 95 percent accuracy, it’s hard to focus on anything other than what the system got wrong. It makes sense because voice assistants are so fragile and even a single misheard word can throw off the usefulness of an answer. For now, Ava is just as accurate as the 3rd-party solutions that are currently powering it, though Duchemin says that once its system gathers more data on a user’s voice it will grow more capable in distinguishing it from the background noise. But for deaf users that often can only achieve 20 percent accuracy in reading lips during a conversation—making up the rest through body language and context—the ability to gain 80 or 90 percent accuracy through using Ava is quite empowering. I reached out to TC contributor and accessibility writer Steven Aquino to hear if this is something the deaf community would actually use. “I grew up in the deaf community, as both of my parents were deaf, so I have that connection between both the deaf and hearing worlds. ASL is my first language,” Aquino said. “I know from experience how difficult is it to have deaf or hard-of-hearing folks involved in conversation. I was always acting as an interpreter when my parents were around hearing family or friends.” Right now, the app allows users to host up to 5 hours of conversation per month. Any user can host a group conversation and participating in another user’s session doesn’t eat up any of your allotted time. For Ava users that move past this 5 hour time frame, they can upgrade to a paid unlimited version of the app for $29 per month. Duchemin detailed that one of the most critical things it getting people on the app and through the setup process as quickly as possible so that groups can get to chatting. The interface of the app is appropriately very simple with just a few controls while much of the screen real estate is devoted to what’s being said. Users can sign into the account with different options, either as someone who’s hard-of-hearing, deaf or hearing, with the app slightly altering the experience to best serve the user’s abilities. The app is on both Android and iOS. Duchemin emphasized that the app definitely isn’t perfect and that there are still a lot of improvements the team is hoping to roll out soon. To them the key was really getting the app out in time for Thanksgiving so that family conversations around the dinner table could be a bit easier and accessible for the 15 million people in the US with hearing issues.
|
Yep, NBCUniversal has invested another $200M in BuzzFeed
|
Anthony Ha
| 2,016
| 11
| 21
|
BuzzFeed just announced that it has raised another $200 million from NBCUniversal. That might sound a little familiar, since the company last year. Plus, that NBCUniversal was doubling down, and the news seemed to be confirmed . The funding release points to past collaborations between the two companies, like joint ad sales and a content partnership between NBC’s Today and BuzzFeed’s food-focused Tasty. Apparently we can expect more of those deals moving forward, with BuzzFeed helping to create and promote native ads for NBCUniversal’s Content Studio, NBCUniversal selling ads on BuzzFeed and BuzzFeed creating “new digital consumer experiences” for NBCUniversal content. “NBCUniversal has been a tremendous partner this past year and we can’t wait to do more with them,” said BuzzFeed CEO Jonah Peretti in a statement. “Our collaboration has allowed us to focus on our respective strengths, learn from each other, and serve our combined audience better with compelling news, entertainment, and advertising offerings that neither company could do on our own. The investment allows us to remain a fully independent company but have access to and resources from the strongest and best media company there is.” Earlier today, BuzzFeed also revealed that it had quietly acquired Scroll, an e-commerce company from the founder of Quirky. So yes, soon it’s also going to be .
|
The Bridge newsletter brings political news to Silicon Valley and tech news to D.C.
|
Anthony Ha
| 2,016
| 11
| 21
|
Tech and politics need a translator — at least according to and , founders of . Corley and Lansing have worked in both worlds — each of them, for example, has served as press secretary for a US senator (Bob Corker and Rob Portman, respectively), while Lansing was also communications director at Politico. They told me they got the idea for The Bridge when Corley moved to San Francisco, and they realized there was an opportunity to create a newsletter that helped people from both industries understand each other — particularly as politics intrudes into tech and vice versa, thanks to issues such as driverless cars. “DC and SF are very similar cities with highly ambitious people on opposite sides of the continent,” Lansing said. “The problem is, they don’t even understand each other. The Bridge is going to fix that.” They compared The Bridge to , which boils down the day’s news into a newsletter aimed at millennial women. Written by Corley and Lansing (at least for now), The Bridge goes out every Tuesday and Thursday morning and includes a rundown of both political and tech news, conveyed in a friendly style and with links to further coverage. Corley said the focus is on helping professionals in both industries understand the “cultural nuances”: “We kind of think of it like meets ” (They also include job postings and event listings.) You can . This might seem to be the perfect time to launch something like The Bridge, thanks to all the discussion around the election — except the election saw the defeat of the candidate , and some writers ( ) have argued that the results represent . “If you’re confused by the results of [the election], you need The Bridge that much more,” Lansing countered. She noted that the newsletter is nonpartisan, and said, “Irrespective of who is in the White House these issues that we mentioned are really coming to a head.”
|
You rarely livestream, but 1M people livechill on Houseparty
|
Josh Constine
| 2,016
| 11
| 21
|
to be a big performance or a one-on-one video call. Just ask the one million people who spend 20 million minutes hanging out each day on group video chat app Houseparty, the latest teen app phenomenon. Built from the ashes of Meerkat, Houseparty is leading the charge for a new kind of synchronous social networking. It’s the closest thing to being with friends in person. Instead of posting to a feed and waiting for Likes, or slinging texts and media back and forth over messaging, Housepartying happens Live. But rather than make one person the center of attention, Houseparty splits the screen up to 8 times so everyone can talk and laugh together. Houseparty’s founder Ben Rubin told me he sees the app as the . But that’s too solitary of a metaphor. The big differentiator from other group video apps like Google Hangouts is that you don’t have to request people to join you, like inviting them to your home. You just get online, and you’re instantly “in the house” partying with any other friends with the app open. That’s why Houseparty is the Internet’s closest version of what sociologists call “ “. The first place is your home, intimate but isolated. The second place is your office, social but formal and full of responsibility. The third place is the coffeehouse, the bar, the mall — a public space for relaxing, socializing, sharing ideas, or goofing off. It can be a psychologically recuperative escape from the rigors of the rest of life. Online, third places often emerge rather than being created intentionally. Inside massively multiplayer online role-playing games like World Of Warcraft, some people prefer to login and see who’s around to hang out with rather than immediately plunging into battle. And virtual worlds like Second Life or Altspace see users dropping in and out of online communities rather than building make-believe homes or consuming content. Houseparty has purposefully created a third place that harnesses the distributed accessibility of the Internet with your real-life social graph mined from your phone book. The result is a way to gather without moving, to simply be social online without the pressure to create. Teens and young adults have plenty of time and a deep desire to congregate with their many friends, but may be disempowered by their lack of self-directed agency, transportation, or home ownership. That’s why Houseparty is taking off with people under the age of 24, who make up over 60% of its users. The app is certainly novel, leading to a huge spike to #3 on the overall in May. But it’s proving to have staying power. Houseparty was back up at #29 this week, and in the few days since talking to the company, it’s now climbed to 1.2 million daily users. The synchronous nature means network effect is everything. Each additional user makes the app significantly more valuable to their friends, so we could see a snowball effect if Houseparty can keep things interesting. And if it can keep the servers online. Houseparty’s parent company Life On Air shrunk its team from 36 down to 11 now after pivoting from its livestreaming app Meerkat. It needed to stretch its $14 million Series B funding from Greylock. Rubin realized Meerkat wasn’t going to beat Periscope and Facebook Live at their own broadcasting game. They had the money and connections to recruit celebrity streamers and focus on premium content output. So Life On Air cut costs and headcount while building Houseparty as an app for normal people on camera, focusing on a frequently used input experience. That’s what will make Houseparty hard for Facebook, Instagram, Twitter, and Snapchat to copy. Houseparty’s synchronous live experience only works if you join the party the second you open the app. It can’t be bolted on as an after thought, the way Instagram tells me it’s thinking about group messaging feature. It’s tough to imagine these giant competitors pivoting so drastically as to put synchronous group video chat front and center. That leaves Houseparty as the place to .
|
How Southeast Asia’s VC scene stacks up to China’s VC scene
|
Romain Dillet
| 2,016
| 11
| 7
|
Some VC firms have a global presence. But startup investment tends to be a very local activity. Kay-Mok Ku, Adrian Li and Jeffrey Paine talked about what it’s like to invest in Southeast Asia at the . While these three firms don’t have the exact same geographic and investment focus, these three VC firms all recently raised new funds, proving that they’re bullish on Southeast Asia. In many ways, Southeast Asia is following in China’s footsteps, but with a ten-year difference. “In Southeast Asia, it’s almost like going back in time,” Kay-Mok Ku said. “Over the last 10 years, there’s been a lot of growth in China. China is obviously a very big market today but it’s also hyper competitive.” That’s why it makes sense to bet on Southeast Asia today if you think that the region is still in the middle of an incredible economic boom. “If you draw that very basic comparison between Indonesia and China in 2008, it looks like there are comparable opportunities,” Adrian Li confirmed. Some startups are even founded by Chinese founders who have moved to the region. “We only have one Chinese LP in our fund. He’s saying Southeast Asia is extremely complex,” Jeffrey Paine said. “It’s fragmented, it’s about the quarter of the size of China. But it’s very easy to expand.” Many Chinese companies are looking at Southeast Asia as an interesting market. Some of them are acquiring local startups, such as Alibaba. Others like Baidu are launching their own products in these markets. “Baidu has had a fairly strong presence in Indonesia with a local team and a completely local product just for Indonesia,” Adrian Li said. “I think they recognize that there’s potential in these local markets.” China is also a specific market with its own rules. Many Chinese companies choose to focus on China, and many startups in Southeast Asia choose not to enter the Chinese market. “China is a very close market and local market today — Southeast Asia is actually much more open,” Kay-Mok Ku said. Adrian Li also agreed, but with some limits. “There are much more compelling reasons for China to go to Southeast Asia than the other way round,” he said. “There are not that many Chinese companies that have reached a critical scale to expand across Southeast Asia. Given all the opportunities in China, making a different strategy to conquer Southeast Asia isn’t a top-of-mind concern for Chinese companies,” he also said later in the discussion. When it comes to the current state of VC investment in Southeast Asia, it’s still quite hard to raise a Series B, C or D round, and the size of the average round is smaller. You can sometimes raise tens of millions of dollars for a seed round in China. It’s not possible in Southeast Asia. “I think it’s also important to distinguish the size of the deals in China and in Southeast Asia. In Indonesia, Series A are between $1.5 million to $3 million,” Adrian Li said. “You get seed deals at a few hundred thousand dollars.” According to Kay Mok Ku, the exit value is still too low as well. That could explain why it’s harder to raise late stage rounds. “In general, raising money is difficult,” Jeffrey Paine said.
|
The Moto Z gets a Nougat update and Daydream certification
|
Brian Heater
| 2,016
| 11
| 21
|
The last time we checked in on (a little over a week ago), the latest version of Google’s operating system was ranked below Gingerbread in terms of adoption. Slowly but surely the needle is starting to move. A handful of manufacturers have announced updates for their devices in the lead up to the holiday, including both variants of the OnePlus 3. This week, Motorola’s modular Moto Z handsets will (not 7.1, mind) this week – or at least the Moto Z and Z Force – no word yet on the Play, but it’ll likely follow fairly shortly after, as that’s how these things tend to go. Along with the software upgrade comes Daydream compatibility, meaning you’ll be able to plug the phone into Google’s new VR headset without any hassle, marking the first non-Google-branded handset to be certified with that official distinction. Perhaps Google is throwing a little love toward its former sub-brand. At the very least, the Pixel is about to get a little less lonely on the Daydream page, and Motorola’s bringing some more key functionality to its most compelling handset.
|
Figuring out what comes after gene sequencing in China
|
Matthew Lynley
| 2,016
| 11
| 7
|
In recent years, a crop of companies have emerged that are using genomic sequencing to detect diseases in order to prevent the negative outcome of those diseases. It’s led to a budding ecosystem of startups that are crunching the data, as well as the cost of that sequencing coming down. But while we’re at the stage of getting the cost of genomic sequencing down and making it more accessible, the applications are the next big question. The biggest potential step is obviously CRISPR and the potential to edit genes — which Illumina GM for Greater China Ruilin Zhao said would be the “$100 billion question” at TechCrunch Disrupt Beijing. For China, that offers a big opportunity because it’s on its way to becoming one of the largest markets for genomic sequencing and testing, he said. “At this stage it’s largely still a concept,” Zhao said. “Having said that, the future is there. We’ve seen successful example. We talk partially about it, a major portion of our company’s work is about detection and finding out. Deciphering the mystery of the gene coding and then trying to think what we can do with it. That’s what CRISPR is trying to do. In the long run, we’re gonna be able to modify genes, but that’s always a guessing game what the timeline is.” Already, there’s a support network blossoming around emerging biotechnology companies — whether that’s through venture capital or otherwise. Illumina, for example, runs an accelerator program that it’s planning to roll out in China soon. That’s led to the emergence of some startups in Silicon Valley like Color Genomics and 23andMe, which have attracted an enormous amount of venture financing. In China, that’s an area that’s also in the works given that it’s on its way to being one of the largest markets in the world for genomic testing, Zhao said. There’s an enormous amount of interest around the globe in gene sequencing and detecting diseases. Even . And part of the reason is that applying machine learning to get a better analysis of the genome and catch potential problems earlier. But one of the biggest problems that’s always going to hound these companies is ensuring that there’s a high degree of accuracy, which as the technology matures, gets better and better, President & CTO at Ardent BioMed David Deng said. Recreational genomic testing will always be an area that is going to have some consumer interest, but there’s also point-of-care testing — essentially getting the results of a test in a matter of minutes — that’ll be an important area for understanding and getting ahead of diseases. As the data science behind it becomes more sophisticated, those answers start to come out faster and faster, making it easier to make more calculated decisions, VGM & Board Secretary of Wondfo Bin Chen said. But with the technology maturing and moving toward the potential of gene editing, there’s always going to be a big question: what happens when we can edit the very basic parts of our lives. And that’s not just a question that’s restricted to Silicon Valley. Or, as Zhao said, “There’ll be a discussion about who we are — are we becoming gods of the whole world?”
|
An investor’s journey to embracing marijuana legalization
|
David Chao
| 2,016
| 11
| 7
|
capitalist and a medical cannabis user. The first time I tried medical cannabis, I had just tried to trigger a chronic condition that causes severe intestinal pain when I eat. I was trying to force my doctor to prescribe painkillers for me. That night, I’d eaten an entire pizza to induce enough pain to get me on drugs. I wound up trying medical marijuana instead and it changed my perspective on the drug. Growing up, I never imagined I would be here. I grew up in Japan and China, where cannabis consumption meant extended jail time. It never occurred to me that using medical marijuana would be an option. Around 2012, I was experiencing flare-ups more regularly and had already spent years meeting with specialists in the Bay Area looking for an effective treatment for my condition. One night I was venting to my friend Blake Krikorian, the co-founder and CEO of SlingMedia, who suggested I try a list of alternative medicine including acupuncture, herbal medicine, and medicinal cannabis. While I respected Blake, I was highly skeptical that cannabis could help. As time passed and my condition got worse; however, I started to conduct more research online. I found that I wasn’t alone with my pain. , . Another NIH estimates that chronic pain affects nearly 100 million Americans, and between 5 to 8 million use opioids for long-term pain management. Personally, I did not want to take my Vicodin prescription given the dulling side effects and risk for addiction. around using cannabis for chronic pain was starting to make sense. I learned that cannabis could be an on everything from migraines to anxiety, is non-habit forming, and doesn’t carry the nasty side effects that some prescription medications have. even found that smoking cannabis reduces pain associated with nerve damage (e.g., neuropathy) in patients with HIV by more than 30 percent compared to placebo. While the concept of using cannabis evoked the image of a pothead smoking out of a giant bong, I found the industry had evolved significantly. Licensed doctors can prescribe marijuana in liquid, edible, and pill forms measured out to an exact dosage just like you might take any other prescription. Gradually I started to warm to the idea of giving it a try. While winning me over took decades, marijuana is perhaps fighting a much tougher fight among my fellow venture capitalists. While our firm invested in , a company in the industry, many Silicon Valley investors still find the business very much taboo. Many VCs are either prohibited from investing in marijuana or aren’t willing to be “that investor”, despite some of them being cannabis advocates or users behind the scenes. That may change soon as the share of Americans who favor legalizing the use of marijuana continues to increase. Pew Research that 57% of U.S. adults say the use of marijuana should be made legal, while 37% say it should be illegal. A decade ago, opinion on legalizing marijuana was nearly the reverse – just 32% favored legalization, while 60% were opposed. Additionally, a recent found that thirteen percent of U.S. adults currently smoke marijuana, nearly double the percentage who reported smoking marijuana only three years ago in 2013. The battleground around cannabis is widespread this election season with cannabis initiatives on the ballot in nine states. California, Arizona, Nevada, Maine and Massachusetts have adult use legalization on the ballot while North Dakota, Arkansas, Montana, and Florida have medical use or expanded medical use on the ballot. In my home state of California, easily the largest cannabis market in the US, medicinal cannabis has been legal for 20 years, but without safeguards or tax revenue. California’s hopes to create a safe, legal, comprehensive system for recreational cannabis use while implementing safeguards, similar to what we’ve implemented for alcohol. There are plenty of positive studies and statistics to support Prop 64 and the benefit of cannabis legalization. In Colorado, where recreational cannabis was implemented in 2014, have found that both the crime rate related to cannabis and the percentage of teen users has fallen. Other benefits include in states that legalize medical marijuana. However, there is even more at stake now. In the year of Black Lives Matter and politicians vehemently expressing the need for criminal justice reform, the legalization of cannabis will reduce the systematic incarceration of African Americans for marijuana possession. Of the for marijuana between 2001 and 2010, 88% were simply for possession, and African Americans were to be arrested in these cases than whites. Prop 64 could be a step towards reforming our criminal justice system by preventing unnecessary incarceration. The night I ate that pizza, I experienced excruciating pain but also had a revelation: After just a small amount of cannabis, my pain was miraculously relieved while my head remained clear — and I discovered that, for me, marijuana was safer and more effective than most painkillers. I believe cannabis should be recreationally legal for adults, and regulated in a similar way to alcohol. My hope is that this November the people of my home state of California agree.
|
Crunch Report | China’s New Cybersecurity Laws
|
Khaled "Tito" Hamze
| 2,016
| 11
| 7
|
Tito Hamze, John Mannes
Tito Hamze
Joe Zolnoski
Joe Zolnoski TechCrunch C/O Tito Hamze
410 Townsend street
Suite 100
San Francisco Ca. 94107
|
GuavaPass, a ClassPass-like service for Asia and the Middle East, raises $5M
|
Jon Russell
| 2,016
| 11
| 7
|
ClassPass is having something of a challenge with the fitness subscription model in the U.S., but that isn’t stopping others elsewhere in the world from trying to make it work , a two-year-old Singapore-based service for finding gyms and fitness centers across Asia and the Middle East, has pulled in $5 million to expand to more markets. Vickers Venture Partners led the round, which GuavaPass said included input from other undisclosed “strategic” investors. ClassPass, the U.S. startup that kicked off the trend for fitness subscription services, has run into tough times lately. , which let customers do as many classes as they wanted for just $180 per month. The finances simply didn’t work — and it isn’t the first time. to make its business more viable, a move that cost it 10 percent of its customer base. That is the U.S., but there are also doubts around this business model for Asia. KFit, the closest rival to GuavaPass, expanded beyond fitness after earlier this year. to move into commerce, in turn throwing up suggestions that it may be looking supplement its revenue streams with a more proven monetization model. Despite question marks around others in its field, GuavaPass, which operates in 10 cities across Asia Pacific and the Middle East, believes that it is developing a sustainable business. “What we’re building in Asia is very unique to Asia, we can tailor the product [for local markets, and] don’t want to create a cookie cutter solution,” Jeffrey Liu, GuavaPass CEO and co-founder told TechCrunch in an interview. “From our viewpoint, we set out to build a sustainable business [and have] always been careful that we weren’t over subsidizing our product,” GuavaPass President and co-founder Rob Pachter added. The duo explained that the grand vision behind GuavaPass is to create a fitness community. Already, the company offers resources on its website, blog posts, newsletters and more, white it recently opened its own 2,000 studio: in Singapore. An “epicenter of fitness options,” Liu and Pachter said the GuavaLabs space is used for branded events, activation, showcasing nutritional products, community events, and more. The idea is to supplement the community building that it is working on in the online space. “We’re now creating this go-to destination that consumes visit on a daily basis,” Liu said. “The longer term plan is to create this sticky community.” “Our objective is to continue our business, it’s only been abut 18 months so we’ve barely scratched the surface. I think we’re on track to build a pretty sustainable business in the long-term,” he added. GuavaPass is looking to use its new financing to expand into new markets. It isn’t being too specific about which new cities it will hit, but Abu Dhabi, Istanbul and Shenzhen are among its confirmed targets. Beyond these expansions, it said the money will be used to “solidify operational infrastructure,” push into the corporate market, and bring on more partners to increase its services.
|
Is LeEco really in trouble?
|
Kate Conger
| 2,016
| 11
| 7
|
Billionaire CEO Yueting Jia told employees in that the company’s rapid growth was leading to stagnation and management problems at , and announced that he would invest $10 million into the company and cut his salary to 15 cents in order to keep LeEco steady as it expands into the United States and develops a self-driving car. But North American operations lead Brian Hui told the audience at today that things aren’t as dire as the letter made them seem. “If you read the letter, it’s not about whether it’s sustainable or not sustainable,” Hui said of LeEco’s growth. “It’s not about running out of money. It’s how you can spend your money wisely… You go through a very aggressive user base growth period before you enter a financially sustainable period. I think this is pretty normal, applying to any kind of startup.” Although the letter led to rumors that LeEco would abandon the development of a self-driving vehicle, Hui said that work on was a “highest priority” and would continue. Hui said that he does not anticipate layoffs on the U.S. team, which has grown from 30 to nearly 600 over the course of the year. The U.S. staff of LeEco is primarily focused on research and development for smart devices and internet ecosystems, he said. “Fast is not about staffing, it’s more about whether you hire the right people,” Hui said. Hui also addressed rumors that suppliers for LeEco’s smartphone business had not been paid, saying that LeEco was not carrying debt to its suppliers. “We will always want to sustain our trust,” he said. “We still maintain a very good relationship with them.” During the U.S. launch of its smartphone , LeEco says all of the phones sold in a little over four hours but declined to say exactly how many units were sold. LeEco launched its first products in the U.S. recently in a big international push. And amid all the chatter of growing too fast, that’s raised questions as to whether LeEco’s move to launch in the U.S. market was the right one. Still despite this, Hui said the company — as it moves on to what he called “phase 2” — was trying to focus on building a more healthy financial operation. So, to that extent, it would seem from Hui that U.S. expansion was already baked into those expectations. [gallery ids="1413150,1413148,1413147"] That launch was accompanied with a in the hopes of placing an exclamation point on the company’s entry into the U.S. Gunning for that kind of publicity, it would seem that the company had very high ambitions for its international expansion. That made the letter sent by Jia to lead to even more scrutiny for the company, given that it hoped to heavily invest in its expansion into highly-competitive markets. For the U.S., it isn’t just about phones and TVs, Hui said. The company is hoping to replicate all of what Hui said were its successes in China. That includes not only its hardware operations, but also content and other ecosystems. Hui said that “smart devices” would be the company’s entry point into the U.S. It’s still going to be an uphill battle across the board, but LeEco is hoping to somehow couple the experiences together using online services, Hui said. “When I was a kid, I used [the Sony] Walkman, but all [Sony’s devices] are silo’d,” Hui said. “If you ambition to have other disruptive services, you want to provide the best service to your customer. And to add value to the customer, you need to have something that couples the entire experience. That’s Internet and cloud.” Jia said in his letter that he would require management to meet a “higher bar” and encourage more collaboration between LeEco’s disparate product teams. “This is the most difficult year in LeEco’s history,” he wrote. “Rumors plunged us into the abyss of hardship and adversity with the stock price of Letv.com falling to rock bottom.” Despite Hui’s optimism, there was a lot to unpack in that letter. And it placed a cloud of uncertainty around the company, which Hui tried to clear out. “I think the most important thing is about not just the letter but the substance in the letter,” he said. “How sincere, how honest, is this company of the thing we are trying to build globally. Between the choices of hiding things, we generally share the challenge we are facing today. The consistent message of the ecosystem we are trying to build, I hope and I believe that if the crowd here or the users pay attention to LeEco, and the users who are using our products and services, you will share and believe that the ecosytem we are trying to build is not just for now but for the future.”
|
Oral arguments for who owns CRISPR-Cas9 start next month
|
Sarah Buhr
| 2,016
| 11
| 7
|
The date has been set for oral arguments to begin over who owns the patent rights to the gene-editing wonder CRISPR-Cas9. Berkeley’s Jennifer Doudna and her colleague from the Max Planck Institute Emmanuelle Charpentier head to court on December 6th to face off against MIT’s Feng Zhang and present their argument before three US Patent and Trademark Office judges on why they, not Zhang, deserve the to own the patent that potentially holds the key to eradicating all inherited diseases. There are several CRISPR patents in existence and MIT owns the lion’s share of them, but Doudna and Charpentier claim they should get the credit for a which Doudna claims is based on her work showing the CRISPR Cas9 system could edit bacterial DNA. However, MIT argues Zhang took it a step further by showing it was also possible to edit the DNA in human cells. Doudna and her team say that’s a logical conclusion, based on her work and not different enough from what she was able to come up with to merit the patent. It’s been a bloody battle, costing tens of millions so far and could go on for years before we know where each side will end up. But the fight isn’t just about getting credit for the work. CRISPR could change entire industries and this particular step holds the potential to wipe out cancers, diabetes, Parkinson’s and many other debilitating but inheritable diseases — and it could be worth billions to whoever owns the rights to license the technology. The order to set these proceedings came down in a four-page letter on November 5 and you can read about them . They will be open to the public and the Alexandria, Virgina-based judges will be able to cross-examine each party and poke holes if they wish through each contention.
|
Is thematic investing the future of venture capital?
|
Harry Stebbings
| 2,016
| 11
| 7
|
Over the last five years, the venture capital ecosystem has seen a significant rise in theme focused investment funds. SaaS funds, hardware funds and my latest guest’s focus, consumer funds have all proliferated. So here to discuss these shifts is Justin Caldbeck, the co-founder of Capital. Thematic funds, in the majority of cases, have the ability to provide a much greater level of operational support, according to Caldbeck. As he suggests, “we have many companies at this stage, in this sector that the resources are very close to hand.” The ability for Caldbeck to advise the Founder on building the onboarding process is optimized. Why? Caldbeck is solely focused on seed stage consumer companies, he only works on this exact use case. His input is not only targeted but researched. He can draw from previously developed pattern recognition of what makes one company successful and the other not. Due to the increasingly crowded VC market, thematic investing is becoming more of a differentiator. Prospective founders witness your domain expertise and gravitate to it, creating the potential for proprietary dealflow. As Caldbeck states, “We want to be the first place for consumer founders to go.” Not only do these founders go to thematic funds but they choose them over other funds. Caldbeck stated that since his move from Lightspeed, he has found it ‘easier to win’. Presenting the case that not only does it get them through the door but also to the signed term sheet. Ultimately, we will continue to see the proliferation of thematic investment funds in the coming years. However, the acknowledgment that such focus can lead to blind spots, in addition to specific verticals getting destroyed by macro environments (Cleantech) mean generalist VC funds will continue to survive and thrive.
|
Casey Neistat’s Beme launches Exit Poll, an app to share who you voted for and why
|
Fitz Tepper
| 2,016
| 11
| 7
|
, the social media app founded by , is getting involved in the election by launching an app called . The concept is simple: You download the app, tap who you are voting for (the options are Hillary or Donald or decline to answer), what state you are in and record a short video explaining your reasoning. Exit Poll then will collate these videos into a Facebook Live stream tomorrow night that will play on its . The idea is you can watch the stream and hear the thoughts and opinions of thousands of Americans participating in the election. While this sounds simple, it actually provides a valuable service. These days it’s harder than ever to drown out the noise and hear what your fellow Americans think. Sure, you can open your Facebook feed and see political posts from your crazy uncle who posts daily links from random right- or left-wing news sites, but that doesn’t tell you anything useful. From the sound of it, Exit Poll will let you hear thoughtful comments from Americans around the country. Of course, this all depends on the quality of the videos submitted by voters. But Neistat explained to me that his company will be screening everything for “profanity and inappropriateness,” which helps increase the odds that the videos shown will actually be interesting and thoughtful. Notably (and despite its name) Exit Poll won’t be releasing any of the aggregate voting data — so there will be no way to see how many people in Pennsylvania using the app voted for Trump or Clinton, even though the startup will have that data. In his launch video, Neistat readily admits that the app will most likely only be useful tomorrow, on Election Day. But the project is part of a bigger move from Beme to experiment with live video, and Facebook Live is obviously the frontrunner in this space. As a refresher, Neistat as a way to share short clips with others around the world. While the app initially soared in popularity, with 1.1 million videos shared in the first week, it had trouble gaining traction and . Beme is still available for download, but Exit Poll may be an indicator that the company is at least experimenting with other ideas. Exit Poll is available for download now and , and this is the where the Live stream will play tomorrow.
|
Apple picked up talent, tech from defunct music startup Omnifone in August
|
Ingrid Lunden
| 2,016
| 11
| 7
|
As the race continues to pick up more subscribers for streaming music services, TechCrunch has learned that one of the most prominent players in the field quietly picked up some talent and tech to advance its position. Apple hired at least 16 employees and purchased select technology from Omnifone, an early player in streaming music services that filed for bankruptcy this summer. The news emerged as Omnifone’s original founder, Rob Lewis (who was no longer with Omnifone in its final years), prepares for his latest streaming music venture, , to launch its first product this week: a music player that plugs into your TV, and a controller that looks a little like a microphone. (More on that .) To be clear, Omnifone is a full acquisition, and it does merit Apple’s standard confirmation. (That confirmation usually goes like this: “Apple buys smaller technology companies from time to time, and we generally do not discuss our purpose or plans.”) There was actually speculation that Apple “ ” or “ ” key assets from Omnifone in July of this year, after a noted that it had found a buyer for parts of the company for $10 million. The rumor at the time was that that buyer was Apple. Those acquisition reports, however, were . But fast forward just one month later, and it looks like at least parts of it were true. About 16 of Omnifone’s London-based all changed their employer to Apple, and a source tells us that in fact there was some technology picked up alongside the talent grab. It includes a license to the patents and right to purchase them in future, but no actual patents. It seems like a logical conclusion that Apple would make the hires and buy the tech from a music startup to work on Apple Music-related services. Music is a fairly new line of business for iPhone maker Apple, but it’s taking on more prominence as the company pushes into new revenue streams to offset slowing smartphone growth. In the last quarter, Apple noted that services revenues collectively were up 24 percent to $6.3 billion from $5 billion a year before. While Apple did not break out Music revenues, it noted that these were up 22 percent. In other words, they grew slightly slower than services overall — one reason why Apple may be looking at ways of improving that with new talent and tech. The list of employees that made the move from Omnifone to Apple includes engineers working on iTunes and apps, among other roles. Omnifone has registered to it, covering areas like digital media identification, streaming and downloading. A source claims that some of the tech has also found its way into Apple Music and iTunes, although Apple declined to comment on any specifics when we asked. Omnifone, founded in 2003 in London, was one of the first companies to try to build a business around unlimited music streaming services for mobile devices — a move that was ahead of its time, coming as it did years before Apple and Android smartphones changed the whole market and when mobile networks were still painfully slow. At its high point, Omnifone had global deals with labels and snagged a number of large companies as clients. Among them were handset makers like Samsung, BlackBerry and Sony Ericsson (later just Sony), as well as large mobile operators who all wanted “sticky” music services to attract more customers, but lacked the expertise or industry connections to build these themselves. Over time, though, those companies backed away from their ambitions, resulting in business losses for Omnifone that it never managed to offset with other activities. Meanwhile, newer, single brands like Spotify emerged as cross-platform leaders in the space — leaving white label provider Omnifone unable to pay its bills and to run up huge debts. Apple has made a number of acquisitions to build out Apple Music. They have included , , and .
|
NXP’s new autonomous driving products focus on safety, self-driving trucks
|
Darrell Etherington
| 2,016
| 11
| 7
|
NXP Semiconductor, which made headlines recently after entering into an agreement to be , unveiled a series of new products today that make it clear why they were a high-value target for the mobile chip giant. Each of three new products NXP announced today promises to advance tech related to self-driving vehicles. In the autonomous car market, NXP revealed a new microcontroller product for use with vehicle radar that will supply four times more computing power to these systems, which will lead to better performance and higher accuracy for radar detection features, which include automatic braking, lane change and cruise assist. NXP also says this will help considerably improve the speed of detection and tracking of objects that might get in an autonomous vehicle’s path, including pedestrians and bikes. The new micro controller is already in testing with some of NXP’s top customers, and will hit wide availability later in 2017. The second new product is a trucking solution developed in partnership with DAF Trucks, Honda and Siemens, which uses vehicle-to-everything (V2X) communication to improve reaction time for connected autonomous truck platoons to achieve 30 times faster reaction time when compared to human-controlled vehicles by 2017. This would allow trucking companies to operate convoys of transport trucks with much higher efficiency when compared to similar arrangements that rely on human reaction time to do things like keep distance, and deal with unexpected events like personal vehicles cutting in between members of the convoy. NXP and Cohda Wireless are teaming up for NXP’s third product announcement. The new partnership will let NXP make use of Cohda’s V2X development algorithms, which will make it a lot easier for customers of both companies to use their technologies together to build in smart decision-making systems that involve culling data from infrastructure like traffic lights, radar detection and other vehicles around them, for instance. As we suspected, it looks like NXP will continue to act as an industry leader in autonomous driving tech under Qualcomm, which should definitely help it return value for its new owner.
|
Tend.ai raises $2M for robot arms that operate multiple 3D printers and workshop machines
|
Devin Coldewey
| 2,016
| 11
| 7
|
Back in June, with its versatile system allowing a robot arm to tend to a dozen or more printers, laser cutters and other modern workshop tools. Now the company has raised $2 million to continue its work, courtesy of . software allows a robot arm to start, stop and monitor multiple 3D printers and other devices, removing finished products and initiating new prints. That means people running businesses don’t have to get up in the middle of the night to check on a long print, or manually unload all the devices in the morning. It’s a big labor saver. “Right now, industrial robots are really good at doing things that need to be done a million times exactly the same,” said co-founder and CEO Mark Silliman in an interview. “But going from enterprise to SMBs [i.e. small and medium businesses] is they’re not necessarily as standardized. So making something that works in a dynamic environment is where a lot of our efforts are going.” https://youtu.be/K5WYLnOcggY To do that, the robot just needs a webcam mounted to its arm, providing a view of its surroundings. The Tend.ai software analyzes what the webcam sees and sends instructions to the arm: hit the start button on machine 6, or turn this dial 30 degrees clockwise on machine 9. The actual motions can be trained by the robot’s owner, but once the basics are down, a shared library of machines and their controls means a bot can get by just knowing, for example, where the power button is, and figure the rest out from there. True Ventures partner Toni Schneider jumped on the opportunity and took the whole $2 million round, indicating no small confidence in the potential here. “It has all the elements we look for,” he said in an interview. “Great team that has worked together before, deep experience, and a really unique, innovative product. I kept talking to people who told me these robots are the future of the industry, but a small business buys one, and then what? This takes something that’s been in use for a while and brings it to a larger user base.” Silliman noted that, since June, the software has been expanded to work with CNC and injection molding machines, which even further opens up the possibilities. He said his early adopter clients will be official customers come January 1, and while he wouldn’t name names, he did say there was “interest from the aerospace industry.” One can imagine how annoying it would be to print 60 slightly different wing designs for testing in a wind tunnel, for instance. Having a robot deal with it means less downtime during printing, and perhaps even fewer devices needed. The funding should help tailor the software to these mysterious but hopefully deep-pocketed potential clients. “At this point [money] is primarily going towards development,” Silliman said. They’re hiring coders and roboticists, but once things get to a certain level of maturity, some will go toward sales, of course. “While we haven’t actually done an investment quite like this before, it did feel a little bit familiar,” Schneider said. “We were investors in MakerBot, which is sort of an iconic 3D printing company, and Glowforge, which is also doing really well. It’s sort of an adjacent space, and it’s a good area for us where we’ve seen success before.”
|
Facebook threatens LinkedIn with job opening features
|
Josh Constine
| 2,016
| 11
| 7
|
Facebook might muscle-in on LinkedIn’s recruiting business with ways for business Pages to promote job listings. The new Jobs features could give companies another reason to drive traffic to their Facebook Page beyond marketing their products in the News Feed, while also allowing them to pay the social network to get their open position in front of more candidates. Today TechCrunch spotted a Jobs tab on its Page, and now Facebook confirms it’s experimenting with a slew of recruiting features. A Facebook spokesperson tells me, “Based on behavior we’ve seen on Facebook, where many small businesses post about their job openings on their Page, we’re running a test for Page admins to create job postings and receive applications from candidates.” The new features could compete with LinkedIn, as well as developers like Work4, Workable and Jobscore that build “Jobs” tab applications that businesses can embed in their Facebook Pages. Perhaps Facebook was prepping for these new features when it tested last year that mimic LinkedIn’s endorsements feature. A new Facebook option in the status update composer allows Pages to formally share a job opening with related details like job title, salary or if it’s full-time versus part-time. The special formatting could differentiate job postings from other content and attract eyeballs amongst the crowded News Feed. These job postings will also show up in a Jobs tab of the Page, creating a dedicated landing place where companies can send job seekers. Businesses would get the added benefit of potentially gaining new followers whether or not someone inquires about the open position, compared to a relatively static Careers page on a company’s website. Job postings will include an “Apply Now” button that launches a standard job application flow, but pre-populated with information from a user’s public profile. That could help people quickly apply for multiple jobs without typing in redundant information. Submitted applications will be received by the Page as a Facebook Message. Though this could bolster Facebook’s push to get businesses using its chat feature, it also might be clumsy for companies to mix job applications in with the customer support requests they typically get via Facebook messages. The potential option to forward job applications to the email address of the company’s recruiter might work better. Businesses will also be able to pay to show their News Feed job postings to more people, directly competing with some of LinkedIn’s ad offerings. Facebook’s opportunity here combines its ubiquitous reach, personal data and engagement. Facebook added Profile Tags last year, which are similar to LinkedIn’s endorsements Pretty much everyone has a Facebook profile, which are often filled with extensive information about their past jobs titles, employers, education and interests. Recruiters can target job posting ads with all this data so they reach people with the right credentials. And because people browse Facebook constantly, those users are likely to eventually see the ads. For comparison, not everyone has a LinkedIn profile. It has 467 million members compared to the 1.79 billion on Facebook. Many LinkedIn users only visit when they’re updating their profile about a new job, or are actively looking for one. But Facebook could reach people not even thinking about a job, yet could be convinced to apply for the chance at a higher salary. That’s the same reason Facebook has been able to build giant brand and performance advertising businesses despite Google’s traditional dominance. You might search on Google when you know what you want to buy, or on LinkedIn if you know you want a job, and their ads can help with demand fulfillment. Facebook offers demand generation, drumming up interest people didn’t know they had. With expected to go through soon, Facebook may be looking to capitalize on the confusion or product stagnation.
|
null |
Darrell Etherington
| 2,016
| 11
| 21
| null |
Review: The NES Classic Edition and all 30 games on it
|
Devin Coldewey
| 2,016
| 11
| 7
|
nostalgia for the holidays this year, like pretty much every year — but the NES Classic Edition, a palm-size recreation of the original console with 30 games built-in, rates highly on the nostalgia scale even for a company whose heart is stuck in the 1980s. It’s already a highly coveted item for millions of 30-something gamers, and make no mistake: This is a love letter to Nintendo’s oldest fans. Now you’re playing with power. First of all, we have to talk about the device itself: It’s tiny. Like, fits in your palm tiny. And while it’s a great reproduction of the original, it’s clear that it’s just for looks. There’s no cartridge slot to put an SD card full of games, no old-school video out on the back, just HDMI. The controller ports aren’t as satisfyingly analog-feeling as the old ones, but that’s really not a big deal. The NES turns on instantly; you’ll be prompted the first time to set up your language, but thereafter you’ll be sent directly to the game selection screen. Enjoy the jaunty NES-style menu theme — I want it for my phone. The menu really shows lovely attention to detail; Nintendo could have phoned it in, but instead took great care, and whole experience is better for it. Hitting Up on the controller brings you to the settings menu, where you’ll find display options (more on this later), language, a couple of miscellaneous tweaks like demo/screen saver mode and auto shutdown, some legal information and a decidedly unhelpful link to download manuals to your phone. Not as familiar looking from this side. That really is one of the big disappointments of the NES Classic for the old-school fan: It would have been satisfying to have the original manual for each game available. Of course, the manuals would have taken up orders of magnitude more space than the games themselves — an NES ROM is on the order of 40-256 kilobytes, while a single manual might be 5-10 megabytes when scanned at a reasonable resolution. Still, it’s a shame. Hitting Down brings you to the Suspend menu, which we’ll come to later. Left and Right navigate through the list of games, and although it isn’t exactly efficient, it’s quick enough you won’t mind not having a more compact view. The Power button, which has the familiar two-step click from the original NES, controls power, obviously. The reset button is used to return to the menu; the game is automatically suspended when you do this. In case you’re wondering, the holes for the screws are in the same places on the back. You can take it apart, though I didn’t risk it. Having played NES games since small times, I know the pretty well, and this gets the feel 95 percent right. The controllers are highly accurate replicas, and while the buttons feel identical, the d-pad seems stiffer — that could simply be because it’s brand new, though. Notably, these controllers are the standard port used on Wii and Wii U, so you can use them on those consoles (we’re guessing this isn’t the case with the ). That said, the cord for the controller is way too short. Well shorter than three feet, which doesn’t help when you want to pass it to the next person on the couch. You can buy extenders, or even a wireless controller, but still, the short cord is a pain. This actually makes it look longer than it is. Believe me, it’s too short. Between this and the necessity of pressing the Reset button on the console itself to save the game, it’s clear Nintendo wants the NES to sit near you on the coffee table or whatnot, but that means running a cord across the living room — not ideal. If you’ve got a nice, tucked-away A/V setup, the NES Classic isn’t going to fit into it. On the plus side, it makes the console very portable. You can unplug it, wrap the cords up and take it to a friend’s house super-easily, and pick up a game where you left off. No cloud saves or accounts here — it’s all on the device itself. I’ve never been a fan of the Virtual Console — I liked the idea, but it never was executed quite right, whether it was the controls, the display style or something else. This time, Nintendo got everything right. Controls are as responsive as they ever were, with no appreciable lag or other weirdness. It’s nice to be playing on the NES-style controller, too. I saw a couple of small graphical glitches in the hours I played, but they were very much the kind you’d see when playing the originals: some graphical corruption that disappears when you walk backwards and forwards again, and the like. Nintendo also has not attempted to improve on the original by doing frame interpolation, removing the 8-sprite-per-line limit or anything like that. This is very much going after the original experience, complete with flickering, well-known bugs in games and so on. Unlike the original NES, of course, this one lets you save your progress in a game at any time. You do this by hitting the Reset button, which puts you back at the menu and shows a winged screenshot floating there, waiting to be resumed, saved or deleted. The interface here is a bit obtuse, though it gets more intuitive with time. Once you hit Reset, you hit down to go to the Suspend menu. There are four slots there, and if they’re empty, you press A to drop the save in there. If a save is already there, you hold A for about half a second and it pushes the previous one out of the way with a cute little animation. The idea is it makes it unlikely you’ll accidentally overwrite another save, but it takes a bit of skill to operate with any speed and confidence. You can also lock saves so they can’t be overwritten by simply pressing Down again while in the Suspend menu. Each game gets its own four slots, which is generally more than enough. One thing to note is that using Suspend overwrites any in-game battery save, like the ones in Zelda or SMB3. To prevent confusion, just use one or the other, or you’ll end up accidentally deleting your game. (Note: Password saves work fine and are fun — well, “fun” — to use.) It’s a bit cumbersome if you use the feature a lot, like saving before each level. You have to hit Restart, then Down, then either A (to save) or Down again (to then select a previous save), then A again. It’s like putting in a cheat code! I accidentally saved or loaded more than once when I meant to do the other, but I suspect muscle memory will eventually take care of that. Presumably Nintendo didn’t want people using this every second, which would be the case if there was a save/load button on the controller itself. I respect that decision, but it’s still kind of annoying. Nintendo was generous with the display modes — when they first announced it, so I’ll just repeat what I said then by way of review: The NES was almost certainly played on a 4:3 CRT television over something like an RF adapter or possibly RCA. The output of the NES, however, was not quite 4:3 (~256x240px), so the pixels would be stretched — that is, not quite square, the way they are on the screen you’re looking at. This, combined with the poor video signal carried by cables at the time and the naturally analog look of CRT phosphors, gave NES games a very distinct and recognizable look. The NES Mini has three display modes (click above for a bigger version): Which you use is really a matter of taste. It might look better to send the pixel-perfect signal and stretch it on your TV rather than in the box. The pixel-perfect and 4:3 modes are bright, colorful and look fantastic — much better than the previous Virtual Console versions. This is a great way to experience these games. Personally, I found the CRT filter to be a bit heavy, darkening the image considerably. Not to my taste, but it’s nice to have the option. People who like this sort of thing will find it the sort of thing they like, and so on. Before I get to the games themselves, let me just say: I think this is a fantastic value. You’re paying $2 per game, plus the well-made hardware and convenience of the save state system. It’s a bargain in a lot of ways. Since controllers are only 10 bucks, it’s easy to recommend buying one first thing. That said, you really should spring for a second controller — many of the games in this first batch are better two-player. $10 isn’t much to ask — but you might consider waiting and buying a wireless or extended-cord version for convenience. I say “this first batch” because it’s inconceivable to me that Nintendo would not put out a few more of these over the next couple of years. The Classic Edition has the early hits and main Nintendo franchises. A Sports Edition comes with Ice Hockey, Double Dribble, Track and Field and all those. The Action Edition comes with original Contra, Rush ‘n Attack, Strider and that stuff. Puzzle edition, RPG edition, etc. It makes too much sense to do this, and too little sense to stop now. Licensing issues may be a problem but Nintendo will make it work if they can sell these games to you for the third or fourth time now. But as those are still hypothetical and this one has such a solid selection, you should go ahead and give them your money as soon as possible. The game selection is a mixed bag — it seems like Nintendo was trying to hit as many types of player as possible without specializing in any one genre. The result is, in my opinion, a few too many arcade games, not enough Contra and an appalling lack of Bionic Commando. Because not everyone was around to play these the first time they came out, here’s a quick rundown of the games you get, what they’re like and which modes they’ve got. (Note: Many of these screenshots are from the Wii U Virtual Console versions; they’ll look much better on the NES Classic Edition.) This Joust-alike was one of the first games to come out on the NES, and it’s pretty basic even for the 8-bit era. It is, however, a joy to control once you get the hang of it. It’s simple enough for kids to play, but watch out, it gets difficult real fast in”balloon trip” mode. : Move fast and you can take out a few guys before they blow up their balloons. Dinosaurs that blow bubbles and turn monsters into fruit and candy? Just go with it. Best played with two players, and suitable for all ages, although, again, difficulty ramps up quickly. If you don’t have the theme song stuck in your head after a few levels, you’re stronger than I. Hold jump to bounce on bubbles. Practice this, it’s critical. One of the all-time classics, Castlevania holds up surprisingly well in terms of art and conception — pay attention and it really seems like you’re infiltrating a haunted castle, not just doing level 1-2 and 3-1 and so on. Focus on mastering the stilted jumping, learn the enemies’ patterns, and don’t be afraid to look online for the ridiculously well-hidden secret hams. Boomerang and holy water are best for bosses. This game may not be as well-remembered as the classic original or groundbreaking third entry, but consult a guide to get you past a few of the more opaque puzzles and I think you’ll find this really is a very innovative and well-crafted action RPG. Plus, how cool is that fire whip? Everyone can beat the first few levels of DK, sure. But then it starts getting pretty hairy. Watch and then see how you match up to world champs. Arcade crossover like this always mean practicing the fundamental movements and situational awareness that, in bygone days, would have saved you a quarter. The levels are designed to trip you up, so pay close attention. The original may be the one we remember, but if we’re honest, it was kind of a clunky game. The sequel is faster, less obtuse and supports two players at the same time in the story. It’s nowhere near as good as River City Ransom, but then again, what is? “Mode B” turns on friendly fire if you want to fight. Fever or Chill? Choose your soundtrack, find a partner and get ready to rage at each other as each fills the other’s flask with bacteria, or viruses or whatever those things are. Like Tetris, this game is extremely easy to learn but difficult to master. One of the all-time NES greats, Excitebike is still a ton of fun. Get a friend, design some outlandish courses and revel in the satisfyingly responsive controls and tricky strategies for winning. Never let up on the turbo. The original, with its many pleasures and many, many pains. RPGs have evolved a lot since the first Final Fantasy, but the fact is all the critical pieces are here: a grandiose story line, stats and gear to obsess over, gold pieces, fantastical creatures and — if you’re smart — a lot of grinding for gold and XP. Remasters on other systems actually improved this game a lot, so unless you want the full experience, seek out one of those. Save often. Especially near wizards. Another arcade port. Galaga is a great game, and this is a good way to hone your skills so you can impress your friends at the arcade (you still go, right?). Practice your timing, remember the bonus wave orders and don’t be afraid to let your ship get captured. Tapping fire is faster and more precise than holding it down. This is one of the games that lent weight to the term “Nintendo Hard.” It’s not as bad as Battletoads, but unlike the arcade Ghost ‘n Goblins, you can’t keep pumping quarters in to keep going. You have to defeat Satan with the lives you get, and if you’re anything like me, you’ll have bite marks on your controller long before that happens. Prepare to die. Gradius still plays extremely well, though it’s far less frantic as modern shmups. This is less about twitch skills and more about knowing and preempting the unique threats posed in every level. It’s still incredibly hard, by the way — I checked. Up, up, down, down, left, right, left, right, b, a, start. Another of the early arcade-style NES games, this one has simple controls that may strike you as restrictive. Play through the first couple of mountains and you’ll see how devious it gets. Don’t play this with anyone you’re not willing to leave behind if they can’t make the jumps. One of my favorite games of all time, Kid Icarus is a long and difficult, but very rewarding, adventure that controls beautifully and hides quite a bit of gameplay depth. Keep a FAQ around to help you out with the hidden scoring mechanisms and pot-dwelling Gods of Poverty in the treasure rooms. Fear the Eggplant Wizard! This was one of the later and most advanced games on the console; inhale your enemies and wield their own power against them. Great graphics and controls, plus smart and cute level design. Don’t let the puffy looks fool you, though, this is a challenging title. Here’s one to pull out to settle grudge matches. You can work together or sabotage each other — just don’t waste that POW block. is inexcusable. This was the correct Mega Man to include: the first was rough around the edges and the ones after this weren’t quite as laser-focused on the core gameplay. It’s got great music, solid controls and it’s just the right level of “Nintendo Hard.” Do Metal Man first and then wreck everything with his weapon. Even himself. It’s truly amazing how advanced the original Metroid was. Not only does it control well and have a huge, labyrinthine map to explore at your own pace, but the music and mood are amazing, too. If you’ve never played through Metroid, you’ve got a treat ahead of you — but be prepared for a serious challenge. Having save states is immensely helpful with this game. Rather than look up a map, get some graph paper and make your own — it’s more fun that way. This game and its sequel really pushed the storytelling on the NES to new heights, with narrative-driven levels and detailed cut scenes in between. I happen to like the sequel better, but the original is great (and punishing). Pay close attention to the actual size and duration of your sword slash. I really don’t have to review Pac-Man, right? I guess it’s worth saying that the NES version is a decent port of the original, though you miss having a joystick. Notice something about the title? Yeah, it isn’t Punch-Out!! The game is exactly the same, but you fight a palette-swapped Tyson at the end — they made him white and changed his name to Mr. Dream. Sad, really, but how often did you even get that far? : Turns out this is a great party game. An underappreciated gem, StarTropics combines interesting action with puzzles and a full RPG overworld and story. Thinking of playing through Zelda or Castlevania again? Why not try this instead? I think you’ll be pleasantly surprised. It’s incomprehensible to me that Nintendo chose to put this one instead of the classic Contra on this thing — a major disappointment, really. But Super C is still a good game, even if it isn’t as universally loved as the previous one. Fire is actually good in this one. This one definitely doesn’t need any introduction. It’s just as good as it always was. Looking for an extra challenge? Watch a few speedruns and see how you stack up. Do yourself a favor and replay this one — the way through. It’s deeply weird, super fun and has great level design that’s totally different from the other Mario games. Probably because it’s just an with a game where you play an Indian family going through a book of stories. Toad is the man on digging levels. Still one of the best games of all time, and always worth playing again. Take advantage of the game save ability and get past World 4 for once! Bring a friend, it’s basically twice the lives and you can learn from each other’s mistakes. Watch “The Wizard.” It may not be as detailed as Madden 2017, but schooling your friends is just as fun. Pit computer players against each other to simulate (and predict) the post season. Never heard of this one. Doesn’t look very good IMHO. So, back in the 1980s they weren’t sure how video game sequels should work, so they basically made the second Zelda a completely different game. It’s not what Zelda fans wanted, but the truth is it’s actually a really good game! The translation is weird and some of the puzzles are random, but with a FAQ handy I think you’ll find Zelda II is a lot of fun if you just pretend it isn’t a Zelda game at all. That shouldn’t be hard, because it really, really doesn’t resemble one. I am Error. Wow! That was a long review. In case you scrolled down here to get the gist (which I actually put above the games), it’s this: It’s a great value, it has a few (though not nearly all) of the best games available for the system and it plays like a dream — apart from some minor gripes regarding the save system. There will probably be more, with different selections, but this is easily good enough to recommend. Buy a second controller (possibly a wireless one) and go to town.
|
Oryx nano antennas allow autonomous cars to see farther, better
|
Kristen Hall-Geisler
| 2,016
| 11
| 7
|
has created a “coherent optical radar system,” according to CEO Rani Wellingstein, that improves the depth perception of autonomous vehicles. The Israel-based company has emerged from stealth after Series A fundraising and is positioning itself as a competitor to lidar systems. Wellingstein pointed out that traditional lidar detects light with photoelectric sensors, “basically measuring the energy of the light.” But Oryx Vision uses what the company describes as nano antennas that are detecting with an electromagnetic wave, which allows the system to access additional information. The advantage is increased range and sensitivity for an autonomous vehicle that needs to know exactly what is surrounding it and what those things are doing. The Oryx antenna operates at a 10-micron wavelength that can see through fog and, unlike lidar, not get blinded by bright sunlight. Oryx engineers began working about six years ago to build a device that could surpass some of lidar’s shortcomings. It took a year and a half just to get the prototype to the point of being a usable, reliable array that can actually be used for imaging. Oryx says its antennas are also less expensive than the lidar systems being used in cutting-edge autonomous vehicles now. The cost of the imager the system uses is about the same as the camera in a mobile phone, and the company is using “the cheapest, most boring laser on earth,” Wellingstein said in a phone interview. The system uses very simple optics, so there’s no steering, no rotating elements and no optical rays to account for. Oryx has 20 people in the company so far, but now that financing is finalized, Oryx plans to hire more talent in Israel, Silicon Valley and Germany, and maybe elsewhere, depending on the demand from auto manufacturers and autonomous vehicle disruptors, Wellingstein said. The company is targeting companies that are building fully autonomous vehicles of the near future rather than the advanced driver assistance systems we’re seeing in cars today. “A lot of the most ambitious players are retreating from these spaces — Apple, Google,” Wellingstein said. “We think the technology aspect is still not there. Autonomous driving is about vision and decision, and there’s currently a gap between the two. There must be another technology, another sensor [besides lidar], to close that gap. Most of the players we’re talking to are not aware there is another technology. That is where we are.”
|
Why it’s so hard to create unbiased artificial intelligence
|
Ben Dickson
| 2,016
| 11
| 7
|
As artificial intelligence and machine learning mature and manifest their potential to take on complicated tasks, we’ve become somewhat expectant that robots can succeed where humans have failed — namely, in putting aside personal biases when making decisions. But as recent cases have shown, like all disruptive technologies, machine learning introduces its own set of unexpected challenges and sometimes yields results that are wrong, unsavory, offensive and not aligned with the moral and ethical standards of human society. While some of these stories might sound amusing, they do lead us to ponder the implications of a future where robots and artificial intelligence take on more critical responsibilities and will have to be held responsible for the possibly wrong decisions they make. At its core, uses algorithms to parse data, extract patterns, learn and make predictions and decisions based on the gleaned insights. This is the mechanics behind many of the technologies we’re seeing every day, such as search engines, face recognition apps and digital assistants. The more data you feed into a machine learning algorithm, the smarter it gets; that’s why every tech firm is looking for ways to gather more data about its customers and users. But at the end of the day, machine learning can only be as smart as the data that’s being fed to it, and therein lies the problem — because the same data used to train machine learning algorithms can teach it to become evil or biased. And like every child, machine learning algorithms tend to pick up the tastes and biases of the persons who rear them. What makes things even more complex is the fact that of the inner workings of their algorithms and treat them as trade secrets. , a machine learning startup, held the world’s . More than 6,000 people submitted their pictures to have their attractiveness evaluated based on factors such as symmetry and wrinkles. This was supposed to relieve us of the social biases that human judges have. But the results turned out to be somewhat disappointing: Of the 44 winners, the majority were white, a few were Asian and only one was dark-skinned. The problem, as , was that the image samples used to train the algorithms weren’t balanced in terms of race and ethnicity. This is not the first time that machine learning’s became an issue. Earlier this year, a language processing algorithm was found to judge white-sounding names such as Emily and Matt to be such as Jamal and Ebony. In another instance, Microsoft was , the chatbot that was designed to mimic the behavior of a teenage girl, after it started spewing offensive tweets. Tay was supposed to ingest comments from users, process them and learn to respond in personalized ways. But it seems that users were more interested in teaching racism and Nazism to Tay. But what happens in more sensitive situations, such as when a person’s life or freedom is at stake? in May found that an algorithm used by law enforcement in Florida gave a higher score to black people when assessing the probability of convicts committing crimes in the future. The list of machine learning fiascos are nearly endless, including a , an ad delivery engine that is and a . Where traditional software is involved, deciding whether an error was due to a user mistake or a design flaw in the software is pretty straightforward. But machine learning is not so transparent, and one of its biggest challenges is determining responsibilities. Developing machine learning software is , and is as much about training the algorithms as it is about writing the software. Even the creators cannot predict with exact precision how the machine will decide, and . Therefore, things got a bit murky when Facebook was accused of being , a component that is at least partly powered by machine learning. And when Republican presidential nominee Donald Trump accused Google of , transparently debunking the claim and explaining the mechanics proved a bit complicated. Things get more sensitive when critical decisions are conferred to artificial intelligence. For instance, if a , who will be held accountable? The driver — or more precisely, the owner — of the car or the developer of the machine learning algorithm? Erasing bias from databases is the key to creating impartial machine learning algorithms. But creating a balanced database is by itself a complicated feat. There is currently no type of regulation or standard governing the data that is used to train machine learning algorithms, and researchers sometimes use and share off-the-shelf frameworks and databases that already have bias ingrained into them. One solution would be to create shared and regulated databases that are in possession of no single entity, thus preventing any party from unilaterally manipulating the data to their own favor. A notable effort in this regard is , a historic partnership between Facebook, Amazon, Google, IBM and Microsoft, some of the leading voices in machine learning innovations, to address some of the fears stemming from the growth of artificial intelligence and machine learning. Part of the goals of the Partnership on AI include addressing the ethical issues of artificial intelligence, and to make sure that a more diverse set of eyes are looking at AI before it reaches the public. Another interesting initiative is , the artificial intelligence company aimed at making AI more transparent and to expand access to it and prevent it from becoming a tool to do evil things. There will probably come a day where robots become intelligent enough to explain their own behavior and correct their mistakes. But we still have a long way to go. Until then, the onus is on us humans to prevent them from amplifying the negative tendencies of the humans who train them. This is something that can be achieved when our efforts are combined, not siloed.
|
Twitter still might save Vine by selling it
|
Josh Constine
| 2,016
| 11
| 7
|
Vine may survive after all. Twitter is currently vetting multiple term sheets from companies offering to buy Vine, and hopes to make a deal soon, multiple sources tell TechCrunch. After announcing its last month, Twitter received a large number of bids, including several from Asia. It’s now working to decide who should run the short-form video app. While TechCrunch couldn’t confirm the names of any of the companies interested in Vine, a rumored bidder was Japanese messaging and gaming company LINE. [Update: We’ve learned that Twitter has narrowed the pool from more than 10 bidders to around 5.] Twitter announced on October 27th that it planned to , but keep the archive of Vines playable and allow users to download their content. But beyond clarifying some of the original announcement in an , it hasn’t made any further announcements about the future of Vine, which makes sense because it’s in active talks to sell the app. One source says that at least some of the offers are for less than $10 million, indicating Twitter might not generate significant revenue directly from selling Vine. However, Vine could still benefit Twitter even if it’s owned by someone who would help it thrive and retain the strong integration between the two apps. Vine content plays instantly in the Twitter stream, bolstering its current parent company’s quest to serve more video that could attract user engagement. If Vine doesn’t shut down and creators keep producing the six-second clips, Twitter could earn money from sponsored content deals arranged by in 2015. [Disclosure: My cousin Darren Lachtman is one of the founders of Niche, but I haven’t spoken to him about today’s news.] Brands who pay creators to promote their products in Vines might also buy ads on Twitter to show those Vines to more users. reported in September that Twitter might look to do something different with Vine. wrote in early October that while Twitter was trying to sell itself, it was also looking to get rid of Vine to cut costs. It later that Vine was costing $10 million a month to run in infrastructure and employees, and Twitter had explored selling it off. After announcing the eventual shut down, selling Vine seemed to be off the table. But TechCrunch has learned that the level of support and mourning for Vine from the internet community boosted the interest of potential acquirers willing to endure the operation costs to own one of the most culturally impactful video platforms. While Vine had been rapidly losing users and star creators, plenty of people wanted to see it live on and were angry Twitter planned to kill off the app. We reached out to Twitter but it declined to comment.
As I wrote last week, . An acquirer could shut down the costly archive, breaking tweets with embedded Vines and pissing off users. Since the Vine brand is closely associated with Twitter, whatever the acquirer does with it could reflect poorly on the former parent company. The sales process could be distracting. Selling for a low price might seem desperate. Someone else could use Vine to compete with Twitter. And if an acquirer revives Vine into a big success, it could make Twitter’s leadership seem incapable. Twitter will likely try to mitigate these risks by choosing the right new owner, perhaps selecting one that agrees to preserve the archive and retain the embed integration. The alternative of keeping Vine alive itself with even a skeleton crew to maintain it could prove too expensive at a time when Twitter is financially struggling. And shutting it down could turn Vine’s stars into evangelists for its competitors. Twitter could refuse the offers. But its best bet is still to let someone else adopt Vine and give it a second chance to grow. Twitter just has to sign on the dotted line.
|
Stealth Uber competitor Zoox valued at $1.5 billion
|
Katie Roof
| 2,016
| 11
| 7
| null |
New Relic extends monitoring to containers and micro service environments
|
Ron Miller
| 2,016
| 11
| 7
|
Monitoring used to be a relatively simple matter. Most companies had a fixed number of applications to monitor. These were usually delivered on the web and lived for a number of years on a fairly fixed number of servers. Today’s environment is far more varied and complex, and made a series of announcements today designed to help customers deal with new ways of delivering applications. In the modern world, companies could be dealing with mobile or web applications. These could live on prem or in the cloud with some delivered in containers as micro services. This all changes how applications performance monitoring services like New Relic approach the craft. To deal with this changing monitoring landscape, the company announced the Digital Intelligence Platform today. The idea is to give customers a broad monitoring dashboard view of the applications under their watch wherever they live and however the customers chooses to deliver them. For some, that bigger view might be divided by job into smaller slices, depending on how a given company divides the monitoring function within an organization. “We have created a single destination for operations and developers to understand the entire app — How the user is managing the app, how the containers are doing, what is the configuration [and so forth],” Bharath Gowda, senior director of product marketing at New Relic told TechCrunch. Containers add another layer of complexity to monitoring applications. They provide a way to deliver an application as a set of discrete micro services. These containers could live for microseconds or a few minutes or more, but their variable and ephemeral nature make them a monitoring challenge. How do you monitor something that doesn’t exist in a static state? To deal with this variability, New Relic has created an entirely new concept called variable tagging, which enables you to tag infrastructure as it’s created in order to track it and see if there are issues with it and how it relates to the rest of the application delivery system. This could allow a company to see a performance issue that is rooted in a micro service, even after it has disappeared. How this will all hold together remains to be seen, but in theory it gives customers much greater control over their applications and infrastructure, regardless of how they are delivering it. This new functionality will be available starting November 16th when it’s officially being announced at , the New Relic customer conference. New Relic went public in December, 2014 after raising over $205 million.
|
Inside an Apple smart home
|
Megan Rose Dickey
| 2,016
| 11
| 7
|
Ever since I saw Disney’s in 1999, I’ve dreamed of a home that cleans itself and cooks breakfast, lunch and dinner. We’re not there yet, but Apple’s HomeKit and gets us (me) a bit closer to that dream. Apple is racing against Amazon and Google to try to get the masses to adopt the smart home. Unlike Amazon and Google, Apple isn’t manufacturing its own devices — with the exception of the Apple TV that acts as a hub. Instead, it’s mostly relying on third-party manufacturers to make their devices compatible with its HomeKit system, and then unite all of those devices through a deep integration with iOS via the Home app. Last week, I had a chance to check out the ultimate Apple HomeKit-wired house, built by home construction company Lennar, in Alameda, Calif. There were over 100 HomeKit devices wired throughout the house, including Lutron Serena smart shades, a Honeywell Lyric Round Wi-Fi thermostat and a Schlage Sense door lock. [gallery ids="1412842,1412843"] Apple’s Home app aims to let you easily control all of your smart, internet-connected products in your home from one app. You can group together certain HomeKit devices to create or edit a new scene. Doing so enables you to activate multiple devices with one command — either a tap of the screen or via Siri. Shown above, you can see your favorite “scenes,” or, the modes you use most often. Scenes can range from what you want your lights to look like when it’s time for dinner to what products you want to have turn on when you come home. The “I’m Home” scene, for example, unlocks your door, turns on the lights and gets the thermostat going. The Apple TV works as a HomeKit hub, which lets you access your home remotely to do things like shut your garage door and view your doorway camera. To ensure complete privacy and security, Apple requires next-level encryption in all HomeKit-certified devices and physical access to the device in order to set it up. My experience with Home was pretty easy, though — Apple had already made sure all of the devices and scenes were set up prior to my arrival. If I were to do this entirely on my own, I wouldn’t install over 100 devices because I don’t own a two-story home. I live in a 3-bedroom apartment that I’m not even on the lease for, so I can’t exactly just do whatever I want with the space. I can, however, do some stuff. Apple pointed me to what I’ve come to call a smart home starter kit. It’s a that lets you control lamps, heaters, fans and other devices from your Home app. To be clear, the iHome smart plug is also compatible with Nest, Amazon Alexa and Smart Things. Peep the video at the top of the page for a glimpse at what it’s like to live in Apple’s version of the smart home.
|
Zipline raises $25 million to deliver medical supplies by drone
|
Lora Kolodny
| 2,016
| 11
| 9
|
has raised $25 million in a Series B funding round to expand its humanitarian delivery drone business in Rwanda, the U.S. and beyond. The startup builds drones and runs delivery services, dropping crucial medical supplies to clinics or hospitals in areas that aren’t accessible by land. Recently, Zipline’s headquarters in Half Moon Bay, California for a behind the scenes look at the company’s fixed-wing drones, launchers and unique landing rigs. led the Series B in Zipline joined by , and said CEO and cofounder Keller Rinaudo. The funding brings the company’s total capital raised to $43 million to-date. Rinaudo said, “We have instant delivery of dinner from companies like Doordash, or groceries from Instacart, so we thought we can do this for medical products.” Zipline recently began a national program to deliver blood on-demand in Rwanda to clinics serving 6 million people there already. The Rwandan government will expand its Zipline service to reach all of the nation’s 11 million citizens in coming months, he said. Zipline will begin dropping off other medical products soon there, including anti-venoms, vaccines and rabies treatments (technically known as rabies post-exposure prophylaxis). The funding will go towards hiring, technology development, and setting up Zipline aerial delivery systems in new markets perhaps Indonesia, Viet Nam and in the U.S., Rinaudo said. Deliveries in the U.S. can only begin with the appropriate regulatory approvals or exemptions, the CEO noted. Even though competing delivery drone makers like or have demonstrated an ability to deliver things like food, or construction supplies as well as drug and other medical supplies with their “birds,” Zipline intends to focus on medical supplies only. In the U.S., that could mean delivering blood or vaccines to rural areas that don’t have ready access to them. But it could also mean delivering an EpiPen on-demand to a suburban home where a kid is having an unexpected, life threatening allergic reaction.
|
Microsoft says it will push for diversity and security from the Trump administration
|
Kate Conger
| 2,016
| 11
| 9
|
Although Donald Trump’s presidential campaign never netted much support from tech companies, at least one company is now congratulating him on his election victory. “Every president-elect deserves our congratulations, best wishes and support for the country as a whole. The peaceful transition of power has been an enduring and vital part of our democracy for over two centuries, and it remains so today,” Microsoft chief legal officer Brad Smith wrote in a . Smith said that Trump’s victory was driven by economic disparity, noting that job growth has not grown evenly for those who have college degrees and those who do not. Trump’s supporters felt “left out and left behind,” he added, noting that Microsoft will work to create new jobs to address the gap. However, Smith’s post also offers a quiet repudiation of Trump’s anti-immigration and anti-encryption stances. “As we think societally about these new opportunities to address those who have been left behind, it’s critically important that we appreciate the continuing national strengths that serve the country so well,” Smith wrote, noting that diversity has a place in strengthening the economy and encouraging job growth. “We’re committed to promoting not just diversity among all the men and women who work here, but the type of inclusive culture that will enable people to do their best work and pursue rewarding careers.” His comments on diversity were echoed at eBay, where CEO Devin Wenig issued a memo to employees reminding them of founder Pierre Omidyar’s immigrant background. “As I travel to our offices around the world, I have the privilege to interact with people of countless nationalities and backgrounds. This diversity is one of the most valuable assets we have as we drive our business forward,” Wenig . Smith also subtly went to bat for consumer privacy and access to encryption (Trump has called for tech companies, most notably Apple, to provide backdoor access to the government). “People will not use technology they do not trust. We’re committed to developing technology that is secure and trusted, both for Americans and for people around the world,” he wrote, referencing Microsoft’s against the Justice Department on these issues. “As we’ve won the cases we’ve brought, we’ve been reminded of one of this country’s greatest strengths, its strong Constitution, independent judiciary, and the overarching rule of law.”
|
Peter Thiel’s Big Gamble
|
Connie Loizos
| 2,016
| 11
| 9
|
Peter Thiel should rightly be admired for sometimes seeing what many others cannot. He understood the power of online money transfers well before most traditional financial institutions. He wrote Facebook a check when it was little more than an interesting startup from another Harvard dropout. Most recently, he anticipated what few of his peers predicted could possibly come to fruition: a Trump presidency. Of course, Thiel more than recognized that Trump’s ascendancy was unstoppable. He spoke out publicly on Trump’s behalf, including in a keynote speech at the Republican National Conventional. He also to Trump’s campaign. If Trump retreated back to the world of reality television and his real estate dealings after this year’s election, as was widely expected to happen, Silicon Valley and the rest of the business world would eventually forget about Thiel’s ringing endorsement of Trump. As distance grew between Trump’s political aspirations and his ability to initiate global annihilation, the tech community would have been increasingly willing to forgive and forget. Instead, Trump is now President-elect of the United States of America. And no matter what your position on that outcome – whether you consider him a genius for acknowledging an angry republic and for to the most powerful office in the world virtually single-handedly, or you deem him a shallow narcissist who has repeatedly women, minorities, civility, and critical thinking — he is wholly inexperienced as a politician. Things will be said that can’t be unsaid. Mistakes will be made. They will also serve as a persistent reminder of Thiel’s support for Trump. You might think it’s for Thiel that Trump isn’t going away, that Thiel will become a powerful and sought-after conduit to the Oval Office. Maybe so. But unless the next president the tech world, it seems just as likely that the extreme opposite is going to prove true, that Trump’s years in office will turn Thiel into the that many in Silicon Valley were eager to cast him as during this divisive campaign season. Thiel made his contrarian bet, and he was right. Now he may discover how deeply unpopular it makes him.
|
Crunch Report | Donald Trump Is 45th President of U.S.
|
Khaled "Tito" Hamze
| 2,016
| 11
| 9
|
Tito Hamze, John Mannes
Tito Hamze
Joe Zolnoski
Joe Zolnoski TechCrunch C/O Tito Hamze
410 Townsend street
Suite 100
San Francisco Ca. 94107
|
LeEco’s Super Bike isn’t electric, but it sure has a lot of gadgets attached
|
Brian Heater
| 2,016
| 11
| 9
|
The LeEco Super Bike may be the perfect analogy for the company that created it. It’s a strange amalgamation that patches together a lot of disparate ideas in a less than elegant way. It’s clearly trying really hard and will no doubt alienate some people with its over-eagerness. Also, weirdly, it’s not electric. That last point was the first thing a rep mentioned to me before looking at the bike earlier today. It’s a strange thing to lead with, but it’s likely already proven a pretty big source of confusion among consumers, so might as well cut that confusion off at the pass, really. It does seem a strange piece of technology to omit from “the world’s first super bike.” At first glance, it brings to mind the bicycle that caused Pee-wee such a headache in the mid-80s. The frame of the thing is a rounded triangle, reminiscent of the Starfleet insignia turned on its side. It gives the bike an added sense of aerodynamics, I suppose, but mostly it just looks weird. There are a trio of big buttons along the top bar; one controls the front-facing light, another snaps the back-wheel parking brake in place and a third takes shots via an optional front-facing camera, and is otherwise vestigial. On the tail of the bike, above the rear wheel, are a pair of lasers that shoot red laser pointer-like lines out the side, creating a theoretical safety barrier to keep cars from getting too close. The lights were particularly effective in the lights of the exhibit space, but will presumably operate much better at night when they really matter. The focal point of the Super Bike’s superness is a four-inch touchscreen that sits up front, between the handlebars, running what the company has deemed “Bike OS” (for obvious reasons) atop Android. The display was actually fairly responsive, and there’s something to be said for built-in GPS, as someone who has gotten lost on his bike several times in the city can certainly attest. Actually using the mapping functionality while riding is another issue entirely. A four-inch screen is still pretty small when you’re trying to find the next turn while cruising down the street. Pretty clear how that could ultimately be a recipe for disaster, even with all the laser lane markers in the world deployed. The display also can be used to do things like selecting music, played through a set of downward-facing on-board speakers. As with most of the company’s non-smartphone/TV offerings, there’s no word yet on pricing or availability here in the States. Though the company assures us that the Super Bike will be available here in some form or other. It’s pretty clear the company mostly wanted to overwhelm its new American audience with all the different things it does in one lavish and bizarre press conference. And really, the Super Bike is the perfect microcosm of that approach.
|
SentiOne raises $3.5 million to monitor social media
|
John Biggs
| 2,016
| 11
| 9
|
While it’s not exactly sentient, . This Polish company watches social media for brands allowing you to create an instant monitor for any topic. Founded by Bartek Baziński, Michael Brzezicki and Kamil Bargiel, the company raised $3.5 million from Venture TFI and Trigon TFI Group to build out their platform internationally. The company has been working on the problem since 2011, but they are now able to generate a sentiment analysis in a few seconds, a feature that few systems offer. They have 1,200 customers and are growing at 5.02 percent. It’s a basic, powerful tool for companies and they seem to be taking to it. “Vast majority of tools on the market, including market leaders do not let you try out and buy their products online, because they’re too complex,” said Bargiel. “We’re the next gen of social listening, we made it so simple and affordable that everyone can start monitoring their brand in 60 seconds.” The company is focused on Europe now and has a net of bots that finds and stores mentions on their own servers. “We’ve built a database of over 12 billion mentions and counting that can be searched in real-time back to 5 years,” he said. The team created their tool when they were attending the Gdansk University of Technology. It was a proof of concept until they realized that these sorts of tools were missing on the market. Using their skills in language processing they were able to test sentiment analysis in 20 languages, a massive improvement over the competition. This means a German company can test sentiment in Spain and vice versa. The team is using the cash to expand further and build out their potentially sentient network of sentiment robots. I, for one, welcome our sentient brand-aware overlords.
|
Hey, Brooklyn, let’s meet up tomorrow
|
John Biggs
| 2,016
| 11
| 9
|
Hello, Brooklyn! I’m restarting my series of micro-meetups in New York with a special tasting at the . The founder, Daric Schlesselman, is a friend of TC and showed off his wares at the last New York Disrupt and he’s offering us a special tasting on Thursday, November 10, 2016 at 7:00 PM. It’s way down in Red Hook, so put on your hiking shoes and let’s see what we can see. This is more of a social gathering, but I’d love to hear your pitches. You can and I hope to see you at the Stillhouse. The address is 6 Bay Street, Brooklyn, NY and the tasting costs $10, payable at the place. We can stroll over to another spot for beers after we’re done.
|
Rigged
|
Sarah Perez
| 2,016
| 11
| 9
|
The Facebook bubble just popped. Half the country today is still in shock. Reality crashed down and many were presented with a world that didn’t match up with the one they’ve inhabited in the months leading up to the U.S. election. As it turns out, that was Facebook’s world. The social media network has become an outsize player in crafting our understanding of the events that take place around us. We’ve known for some time that in terms of exposing us to differing viewpoints. But only today are some realizing how powerful its influence has become. On Facebook, people were told the world was either a disaster, or seeing monumental progress. On Facebook, a Trump victory was likely, or a Clinton win was all but assured. On Facebook, the thoughts in your head turned into news articles you liked, turned into things you could share. On Facebook, . And the louder we screamed, the more our time on the site increased. . Facebook didn’t just reflect your views back to you. It magnified and distorted them through the lens of sensational and often falsified stories. And it got away with it by throwing up its hands and claiming, “ Rather, it pretends to be a neutral platform where people can share whatever they like, within reason. Its teams of moderators police the site for content like pornography or the illegal sales of firearms or drugs, and other generally prohibited things. But beyond that, it turns a blind eye to the nature of the content within its walls. of fake news sites with have filled the network even as Facebook abdicated responsibility for the disinformation it lets virally spread. It even went so far as to news editors who managed the Trends section, leaving the matter up to an impartial, but entirely fallible, algorithm. This wholesale elimination of human judgement from the site’s news machinery could not have come at a worse time for the election. The algorithm later trended a number of stories that were “profoundly inaccurate,” according to a that tracked the occurrences of fake news in this high-profile section of Facebook’s platform. Facebook showed users a story saying 9/11 was an inside job, a false that Fox News anchor Megyn Kelly was fired, a story about how someone praying was kicked off a college campus. It even promoted a story about iPhone that works like an Aladdin’s lamp from a whose name is “FakingNews.” Facebook brushed asides these offenses as mistakes, claiming it would do better in the future. But Facebook’s focus has been on making it easier for publishers to share on its network, not vetting their content. It invested in like Instant Articles which make news reading more painless with quick-loading pages free from burdensome scripts and ads. It works to figure out how to keep users on site for longer, so they can click on ever more personalized, targeted ads. Of course, one way to increase engagement is to make people feel good when they arrive. And Facebook knows how to control your feelings, because it has studied this extensively. The company where it manipulated the posts on 689,000 users’ home pages to see if it could make them feel more positive or negative emotions. Turns out, it can. People at the time said that Facebook could use the data it collected to figure out how to feed us a stream of happy thoughts to keep us on its site. Sound familiar? It should. The WSJ documented this in the realm of politics with its . In other words, Facebook spoon feeds us what we want to hear, while minimizing our exposure to the opposing viewpoint. The results have been beneficial for Facebook, to say the least. The network now has 1.79 billion monthly active users as of September, 2016. In its last , it pulled in another $7 billion in revenue: $2.379 billion of that was profit, up 16 percent over the $2.05 billion it brought in last quarter, and up 160 percent year over year. The problem of the Facebook bubble matters not just because we’ve been duped by the algorithms, but the significant role Facebook plays in the dissemination of news. [gallery ids="1413932,1413933"] Today, a majority (62 percent) of U.S. adults get news on social media, which includes Facebook and other sites, a Pew Research study from May 2016 . And Facebook is the largest social networking site, reaching 67 percent of U.S. adults. Two-thirds of Facebook users (66 percent) get their news on the site — a figure that amounts to 44 percent of the general population, from Pew Research. That’s in 2014. To make matters worse, social media is a poor platform for getting people to understand opposing views. Another that only 20 percent of users modified their stance on a social or political issue because of what they saw on social media. A smaller 17 percent said they changed their views on a political candidate because of this. When Pew then examined those changes in more detail, it found that social media had often pointed people in a more negative direction. That is, people who changed their minds on Clinton were more than three times as likely to have gone negative on her, and people who changed their minds on Trump were nearly five times as likely to have gone negative on him. In addition, 82 percent of social media users said they changed their minds on a candidate and 79 percent never changed their minds on a social or political issue because of social media. So in terms of convincing anyone of anything, Facebook wasn’t the place to do that. We’ve known this about Facebook for some time, but many never felt it quite as profoundly as today. A personalized feed that tells you what you want to hear is great… until it’s not. Last night and into this morning, people began to realize their sources had bad information; their data was wrong and the polls were off. And, most importantly, they discovered their crazy uncle wasn’t an outlier — he represented half of a very disturbed, very angry nation.
|
Here are the tech policies Trump promised to implement as president
|
Brian Heater
| 2,016
| 11
| 9
|
Trump’s not a tech guy, that much we know. He reportedly doesn’t have much time for a . During a CNN town hall moderated by Anderson Cooper, that he largely composes tweets by, “shout it out to one of the young ladies who are tremendous. I’ll just shout it out, and they’ll do it.” But while he’ll never be an early adopter (and he may or may not actually use a computer), the president-elect’s policies will have a wide-ranging impact on technology, which in turn will continue to play a larger role in our personal and professional lives. While technology was never a centerpiece to Trump’s campaign, the candidate has touched upon the space from time to time during interviews, speeches and debates. What follows are some of what we can perhaps expect from a Trump presidency. Trump has had some pretty tough words for Cupertino. Five days after Valentine’s Day, he called for a boycott of the company’s products over its encryption stance following the San Bernardino mass shooting and bombing attempt. A month prior, the candidate harnessed his distrust of China to announce his plans to quite literally its products in the United States. “We’re going to get Apple to build their damn computers and things in this country instead of in other countries,” he told a crowd at the University of Virginia. The candidate also made promises about Ford cars and Oreo cookies. The concept of global warming was created by and for the Chinese in order to make U.S. manufacturing non-competitive. — Donald J. Trump (@realDonaldTrump) In interviews, Trump had a lot to say about the matter. Mostly involving the fact that he doesn’t believe it exists. Or that it’s not man-made. Or if it is man-made, it’s part of a man-made hoax perpetrated by China. Speaking with CNN in September, , “I believe in clean air. Immaculate. But I don’t believe in climate change.” In a tweet from 2012, Trump stated, “The concept of global warming was created by and for the Chinese in order to make U.S. manufacturing non-competitive.” A few years later, when complaining in earnest, he would write the statement off as a “joke,” but continued to call the phenomenon a hoax, stating that the U.S. ought to stop spending money on it. Here’s one such statement from a December 2015 rally. “Obama’s talking about all of this with the global warming and … a lot of it’s a hoax. It’s a hoax. I mean, it’s a money-making industry, okay? It’s a hoax, a lot of it.” Seems pretty clear. He has also cited cold weather as proof of the aforementioned H-word. “If I become president, do they have problems,” Trump said back in February, referring to Jeff Bezos and Amazon. “They’re going to have such problems.” At the time, the candidate was taking specific issue with this fellow billionaire’s purchase of . He’s accused Bezos of using the as both a tax shelter and a method for swaying political influence to benefit himself and Amazon. Finally trashed by . Will still reserve him a seat on the Blue Origin rocket. — Jeff Bezos (@JeffBezos) For his part, Bezos reserved Trump a special (one assumes one-way) ticket on one of his Blue Origin rockets. Speaking of space, Trump says he wants to make it great again. The candidate in NASA’s home turf of Florida that he would, “free NASA from the restriction of serving primarily as a logistical agency for low Earth-orbit activities. We will instead refocus on space exploration. Under a Trump administration, Florida and America will lead the way into the stars.” But while Trump has expressed interest in going to infinity and beyond, his NASA plans, predictably, don’t offer much in the way of self-examination. Which is to say that, under him, the agency won’t spend much time looking at life on Earth. In published last month, a pair of advisors stated that, “NASA should be focused primarily on deep-space activities rather than Earth-centric.” The view is due likely in no small part to its ties to climate change, a phenomenon the president-elect does not believe to be tied to man-made activity. But hey, as long as we find a new planet to live on before we completely destroy this one, we should be good, right? Obama’s attack on the internet is another top down power grab. Net neutrality is the Fairness Doctrine. Will target conservative media. — Donald J. Trump (@realDonaldTrump) Trump’s major gripe with net neutrality seems to involve his equating the concept with censorship. He’s referred to it a “top down power grab” and compared it unfavorably to the FCC’s fairness doctrine, which attempted to require broadcasters to give equal time to all sides of an issue. His planned appointment of anti-regulatory crusader Jeffrey Eisenach has been viewed as problematic by net neutrality supporters. The cybersecurity policy Trump laid out at his first debate with Hillary Clinton was… . Here’s part of his response when asked about online attacks by moderator Lester Holt, which failed to implicate Russia, [W]e had to get very, very tough on cyber and cyber warfare. It is a huge problem. I have a son—he’s 10 years old. He has computers. He is so good with these computers. It’s unbelievable. The security aspect of cyber is very, very tough. And maybe, it’s hardly doable. But I will say, we are not doing the job we should be doing. Trump’s site than his own comments (which, admittedly, isn’t a hard thing to do), repeatedly hitting Clinton on her ongoing email woes, citing various issues with hacking over the years and ordering a review of the U.S.’s online vulnerabilities.
|
Adam Bain departs Twitter, Anthony Noto will take over as COO
|
Anthony Ha
| 2,016
| 11
| 9
|
Twitter just announced that Chief Operating Officer Adam Bain is departing. Bain has been with Twitter since 2010, building the company’s ad business and at the same time that founder Jack Dorsey was named CEO. Bain that he’s now “ready to change gears and do something new outside the company,” while , “I look at what we’ve built together and how far we’ve come, and I feel like I can leave the company now in a place of strength.” He didn’t say what he’s doing next. I’m excited to take w me everything I’ve experienced at Twitter & do something totally different and new outside the company. Stay tuned! — adam bain (@adambain) Anthony Noto, currently Twitter’s chief financial officer, will replace Bain as COO, according to . Noto will continue to serve as CFO, as well, until the company finds someone new for the role. This shift comes at a tumultuous time for Twitter, with the company last month after failing to find an acquirer. Twitter stock actually popped this morning following , though as I write this, it’s down 3 percent in after-hours trading.
|
A serial founder on Trump: “He’s a startup that disrupted the establishment”
|
Connie Loizos
| 2,016
| 11
| 9
|
, a longtime New Yorker, has helped create hundreds of jobs for Americans and others. Among the companies he has started is , one of the largest free classifieds sites in the world — one that was acquired over time by the African conglomerate Naspers for $250 million. Grinda more recently co-founded , the peer-to-peer used car marketplace based in California; , a New York-based platform that buys high-end luxury bags from their owners for cash; and , a Sao Paulo, Brazil-based car-buying service that will buy individuals’ cars for cash in an hour’s time. Grinda and longtime business partner, Jose Marin, also plug between $15 million and $20 million of their own capital into startups each year through their joint vehicle, . But though he sounds it, Grinda isn’t American. He doesn’t have dual citizenship. He’s “pure French.” He just happened to head to Princeton at age 17, and he hasn’t much wanted to leave the East Coast since. So what does this European make of a new U.S. president who has Silicon Valley on edge? Because he’s a global operator and because he doesn’t live in the Bay Area, we talked with Grinda earlier today about President-elect Trump and whether he’s concerned about what comes next. FG: Public market investors, limited partners in venture funds and private equity firms — they don’t like uncertainty. What they don’t know is the actual set of policies coming down the line that could impact them going forward. What will be his tax policy? What will his administration regulate and deregulate? It’s not like [Trump’s team] came forward with a well-thought-out set of policy proposals. It was all kind of vacuous. So I think investors will be more cautious until they understand what a Trump presidency means. FG: I don’t spend much time thinking about politics. I’m not sure it has a real impact on day-to-day life. It’s a large part of the reason I’m on the internet. I like its deregulated, fast-moving nature. FG: There’s no hard data regarding what is going to be done. My only concerns are around the uncertainty. FC: I dislike the guy. I dislike populism and most of the things he said and much of what he stands for. I’m pro immigration and probably more socially liberally than anyone I know. But look, he’s a startup who has disrupted the establishment. He used a lot of the same tactics that a startup would use to get free press, frankly. He created a story that was compelling enough that he garnered press all the time and so had much lower acquisition costs than the other candidates. Jeb Bush was paying something like $5,000 per voter in the GOP primaries, where Trump was paying about $300. In startup terms, he had an effective distribution and a marketing strategy and messaging that people found compelling. I think he proved the adage that any press is good press. And the establishment only realized this was dangerous once it was too late. By the way, I think the same is true of jihadists; all the media attention that ISIS receives makes it easier for them to get recruits. FG: Clearly, there’s a percentage of the population that’s been left behind and not listened to and we need to find a better way to deal with that. I am an optimist. I do believe the tide of history is toward more liberalism and the quality of life improving. Sometimes, you have pushback, but in the grand scheme of things, it doesn’t seem to matter. As horrible as the Great Depression must have been to live through, it barely registers in the bigger picture. I think the next four years will be a blip, too. Has globalization had some losers? Absolutely. If you’re a high school dropout, your relative job position in the job market hasn’t been great over the last 30 years. It’s a class of people who haven’t been heard, and we haven’t been good at retraining or them or integrating them into the success of this country. We need to focus on opportunities to refocus the education system and retrain them and that’s the message that’s been sent and maybe it’ll force us to get our act together. FG: I think there are things [Trump] could do well. The U.S. hasn’t had a good infrastructure program for years. To create jobs, the easiest way isn’t to block technology but to build better roads and bridges and airports and the things that are needed and create lots of jobs. The reality, too, is that we’re crumbling under red tape. We have an outrageously ineffective corporate and personal tax system that is run inefficiently. And we’re under a mountain of regulations. If you’re in the offline world for example, the burden for construction alone is limited to an insane degree by . If he could do these things, we might be able to make the best of a bad situation. I know I’d feel better.
|
Trump promises to bring back manufacturing jobs, but robots won’t let him
|
Lora Kolodny
| 2,016
| 11
| 9
|
For Americans struggling with stagnant wages, under- or un-employment, one of Donald Trump’s most appealing campaign promises was to to the U.S. Navigating the complexities of policy, tariffs and geopolitics would make that hard enough already for the president elect. But technology will make this promise nearly impossible to fulfill. Why? Because manufacturing jobs are increasingly done by robots, not people. Robotics have already helped reduce reliance on labor overseas for manufacturers in automotive, electrical and electronics industries, according to a fresh policy report from the And automation does allow manufacturers to do business in the U.S. when they may have chosen to do it in Southeast Asia or somewhere else, before. But when manufacturing returns to the states, jobs aren’t coming with it in high numbers. Automation has left workers in developing nations without employment, the report notes, and the U.S. faces the same prospect. Startups creating interesting robotics that stand to “steal” jobs from people, either here or abroad, are attracting seed and venture funding. Players in this space include burger flipping and pizza making robots, respectively, from and , painting robots from . Then there are the likes of and robots configurable for a wide range of purposes in manufacturing and elsewhere. It’s not just startups, though. Large brands like and have shed contractors and embraced robotics and 3-D printing to make their shoes. in the field, and major companies like and rely heavily on robots for logistics and warehousing. UNCTAD’s report explained, “increased use of robots in developed countries erodes traditional labor-cost advantage of developing countries,” and is already having a global impact. It’s not just that robots ramp up productivity. They can help companies streamline all kinds of headaches affiliated with setting up shop overseas. A manufacturer outsourcing to vendors or hiring employees across borders will have to deal with the costs of managing people, design, quality, safety, customs and logistics, regulatory compliance and intellectual property from afar. “Reshoring” becomes more appealing with every technological advance in robots, since it reduces administrative and legal overhead as well as labor costs. And robots aren’t getting dumber, obviously. Advances in computer vision and artificial intelligence promise to make robots, and the software-brains inside of them, even more competitive with people, especially in manufacturing but even in physical security or hospitality. Yes, robot , and are a thing. Let’s just hope there’s always a market for “handmade” goods and human-delivered services, and perhaps a robot that can help teach former employees new work skills.
|
Bikeshare conference brings pedal power to the people
|
Kristen Hall-Geisler
| 2,016
| 11
| 9
|
As the begins its annual conference in Austin, Texas, participants will be focusing on social equity. If you’re wondering what that means, wonder no more, thanks to a statement from the NABSA’s project manager, Matt Martin: “Social equity in bikeshare means making a healthy and versatile form of transportation accessible to those who really need it, from students on a budget to mothers in low-income communities.” I also spoke with the organization’s president at length about bikesharing and how it fits into a changing transportation landscape. “Many people that use bikeshare are people commuting in multiple ways,” she said. “They take the train and bus and then hop on a bike in the end. It’s so convenient and cost effective, and people are using it for practical reasons.” Bikesharing has been popular in European countries for decades, but now more North American cities are seeing racks of shared bikes at busy intersections. There are with bikeshare programs, including Portland’s new Nike-sponsored system, Ford’s sponsorship of availability in San Francisco, and Washington DC’s Capital Bikeshare. “It’s proven successful in almost 100 cities at this point,” Freedman said. “We hear a lot about large cities’ [programs], like New York and Chicago. But the reality is many midsized cities have run successful bikeshares.” There were early attempts at Euro-style bikesharing, but many times bikes were stolen or trashed rather than shared nicely. The difference in these latest systems, Freedman said, is the thing that’s making a difference in almost every aspect of transportation: technology. “What made bikeshare succeed is the same as with Uber and Lyft. There’s a software system with a hardware component and sophisticated software on the back end. You can’t steal a bike; you need a credit card to get access. But technology makes it so easy – just swipe a key or card and you have a bike. With the Portland system, you can use your cell phone.” She noted that the success of a bikeshare program is different in every city, and even in each community being served. “Is the system reaching people regardless of race, income, and nationality, or is it just serving a slice of the market? As you expand your service area outside the core of the city, you’re going to do that for equity reasons. It’s public transportation. Outside the core, there will be fewer riders per station and per bike, but it’s a public service for the public good.” Freedman predicts that in the next couple of decades, most people will not be driving, nor will they own their transportation. “People will be mode agnostic,” she went on. “What an individual cares about is getting from point A to point B the fastest way possible, the least expensive way possible, with the most comfortable experience.” Users will mix bikeshare, walking, public transit, and rideshare every week — and maybe even every day. “You’ll see all of the modes intertwine,” Freedman said. “And that will be made possible by better tech that allows a single payment for all systems.”
|
LeEco offers a budget take on Gear VR
|
Brian Heater
| 2,016
| 11
| 9
|
LeEco is a weird company and the ExploreVR is a weird headset. It’s essentially a Gear VR competitor that promises to deliver a low-end virtual reality experience for a fraction of the price of Samsung’s peripheral. It is, of course, not the company’s first VR headset. In typical LeEco fashion, the company’s already got a bunch on the market already – because when LeEco does something, it does a lot of it. Most compelling of its current offerings in the space is a standalone VR headset that runs about $200, removing the handset from the equation entirely. But, while someone I spoke with at an event for the company assured me that that model provides the best experience of all of the offerings LeEco has in China, there’s no word when or if such a thing might arrive here in the States. The ExploreVR, on the other hand, is coming, but when and for exactly how much have yet to be determined – though it will apparently be well under the Gear’s $99 asking price. From a design standpoint, it looks a fair bit like the Gear. There’s a touch control pad on the side, a USB-C port for charging on the bottom and a slot in the front for the phone (in this case the LeMax 2 Pro), with a reflective front face that snaps over. The most notable aesthetic difference is the swapping of the headband for a large plastic curve that rests of top of the head, weighted in the back to keep it in place. It’s not as adjustable as the headband, and it’s more likely to muss your hair than Samsung’s offering. But it’s certainly different. So that’s something. The headset fits reasonably well and there’s enough room inside to use it with a pair of glasses. Since there’s no band to hold it in place, it will end up resting on your nose. The rep I spoke with added that there will be some aesthetic tweaks to the hardware before it hits the US, but refused to go into detail. The device uses LeEco’s own content store (not up and running on these units) with mostly Chinese language VR video. Interestingly, the company also plans to open the its content store up to 3D party hardware, like the Rift, Vive and Gear. The demo I got featured two short videos – 3D guided trips through classic paintings. The motion was a bit jerky and there was some clear distortion around the borders of the lenses, which doesn’t bode well for the ultimate experience, given that these were fairly controlled demos. Depending on how much the hardware ultimately costs, the sheer price point may make it tempting for some viewers, but the company is about to get an awful lot more completion through Daydream headsets (LeEco said it has no plan to develop for that platform at the moment). Of course, there’s a built-in audience of those who buy LeEco handsets (ExploreVR will apparently be compatible with all of the phones the company sells in the States), and if the company can build a large enough English language content library, there could be some appeal there, as well.
|
This Indian team plans to be the first private company to land on the moon
|
Emily Calandrelli
| 2,016
| 11
| 9
|
, a private company located in Bangalore, India, is the one and only Indian entrant into the , the international competition to land a private spacecraft on the moon. For TeamIndus, there are a few, exciting things at stake. If they’re first in the competition, they’ll win $20 million and become the first private company to land on the moon. If they’re successful at all, they’ll achieve something their own government’s space agency has yet to accomplish: soft-landing on the moon. The company had bit of a late start in the competition, joining three years after the Google Lunar X-Prize was announced. But today, TeamIndus employs over 100 people, was one of three teams to win X-Prize’s $1 million for lunar lander technology, and is on their way to raising upwards of $10 million. These factors have helped secured TeamIndus’s spot as a lead contender in the contest, but they certainly have some ways to go. The company has yet to secure a launch contract, for example, which is something they’ll need to do soon if they want to meet the X-Prize’s deadline of landing on the moon by December 2017. Their mission plan involves launching a spacecraft on an Indian rocket into low Earth orbit. The spacecraft will complete two orbits around our planet before initiating a propulsive maneuver that will propel it toward the moon. An insertion burn will place the spacecraft into lunar orbit where it will travel around the moon three to four times before beginning the most complicated part of its mission: descent to the surface. Once the spacecraft touches down, the TeamIndus rover will deploy from the parent spacecraft. To complete the X-Prize requirements, the lander will roll at least 500 meters along the surface and send high definition images and video of the moon back to Earth. Illustration of TeamIndus Rover on the Moon / Image courtesy of TeamIndus In addition to a rover, TeamIndus will bring a student science experiment, selected from more than 3,000 possible experiments submitted to their competition, to the lunar surface. When asked what the delay was in securing a launch, TeamIndus’ co-founder, India’s Polar Satellite Launch Vehicle / Image courtesy of ISRO Rahul Narayan explained that their situation is a bit different than other X-Prize competitors. While other teams have signed launch contracts with private rocket companies, TeamIndus will be flying with a government owned and operated rocket. Specifically, TeamIndus will ride on the Indian Space Research Organization’s , a vehicle with a solid history of reliability. In June, PSLV its 35 consecutive successful mission. By purchasing a ride on PSLV, TeamIndus is setting a precedent for doing business with the Indian government. Narayan noted that because this is a first-of-its-kind situation, it is taking a bit more time to sort everything out. “ISRO has never sold a dedicated launch vehicle to a private company before. We’re breaking new ground here.” Rahul Narayan, TeamIndus co-founder When asked if they would need permission from their government to go to the moon (as ), Narayan explained that because PSLV is owned by the Indian government, a signed launch contract would inherently include their government’s permission. Compared to the other rockets selected to bring the X-Prize teams to the moon, the PSLV is arguably the most reliable and may be a great advantage for TeamIndus. , an Israeli team, has secured a launch on a SpaceX Falcon 9, which has had two mission failures in the past 18 months causing unavoidable schedule delays. Moon Express has in Rocket Lab’s , a vehicle from a new New Zealand-based launch provider that has yet to fly. If all else is equal, a team’s choice of rocket may determine who gets to the moon first. Of course, for their choice of rocket to matter, TeamIndus must first be able to afford the ride. In accordance with the Google Lunar X-Prize , which state that 90% of each team’s mission costs must be privately funded, TeamIndus cannot be “gifted” a ride to the moon from the Indian government. ISRO, however, will help TeamIndus in other ways. Narayan told TechCrunch that TeamIndus will have access to ISRO’s vibration and acoustic test facilities in order to qualify their spacecraft for flight. That will certainly save some money, but that type of assistance is with other X-prize competitors. All other expenses, including the cost of the rocket, ground operations, and technology development, will add up to about $70 million to $80 million, Narayan estimated. To date, the company has publicly stated that they’ve raised about $3 million ($500,000 in seed funding, $1.5 million in Series A, $1 million from the X-PRIZE milestone award). But Narayan told TechCrunch that TeamIndus is closing another round soon that will bring their total up to between $10 and $15 million and plans to add more people to their 100-person team. Putting those numbers in perspective, , a Google Lunar X-PRIZE team from the U.S. has raised over $30 million and as of August only had about 25 employees. TeamIndus spacecraft (the rover will be carried inside) / Image courtesy of TeamIndus While employee salaries are only one piece of the overall budget, the for engineers in India is significantly lower than engineering salaries in the United States, so a company like TeamIndus can afford to build a bigger team. Narayan stated that TeamIndus’ ability to hire more people was an asset in a competition as complex as the Google Lunar X-PRIZE. While a team of that size is unique in this competition, we’ll have to wait and see if the ability to hire more engineers is truly an advantage. So what happens after the contest is over? Teams like Moon Express and , another U.S. based competitor, have stated their intent to sustain a lunar-based business after the prize has been won. TeamIndus, however, has a different plan. Narayan said that while they don’t expect to continue with lunar-based missions after the contest, they do hope to maintain business in the space industry. He noted that TeamIndus has built up a capability for designing, prototyping, and analyzing space-grade technology that could be used to create a business building satellites. But for now, their focus remains on the moon. Narayan, along with his 100 employees, are all working toward the goal of bringing an Indian to the moon for the first time, something that he spoke about with great pride. When asked how it would feel if TeamIndus were to win the Google Lunar X-Prize, Narayan said “It would be the same feeling you’d have if your team won the Cricket World Cup.”
|
null |
Greg Kumparak
| 2,016
| 11
| 7
| null |
Watch Acceleprise’s Demo Day here
|
Samantha O'Keefe
| 2,016
| 11
| 9
|
TechCrunch is pleased to bring you Acceleprise’s Cohort 5 Demo Day, today, November 9th, starting at 10:00am PT. – HR and payroll platform for developing countries finance providers accurately price capital, and monitor their portfolio for risks and opportunities. – Real time visual collaboration integrated with 3D models for construction companies – A personalized mobile banking platform for millennials – A complete business and vendor management platform for the fitness industry – The digital platform for commercial insurance brokers
|
Google ‘strongly’ recommends against third-party fast charging technology on Android
|
Brian Heater
| 2,016
| 11
| 9
|
Everyone, it seems, is going the fast charging route these days. Thanks to the nearing ubiquity of USB-C on flagship devices, the feature is quickly becoming a standard – “standard” in the sense that everyone is doing it, not so much that there’s any consistency to the tech. All in all, it’s a nice addition to manufacturers’ newfound focus on battery life. And while Google has embraced its own version on its new Pixel devices, the company’s not so keen on letting everyone implement their own version. In newly released papers issued for Nougat, the company stops short of an outright ban on technologies like Qualcomm’s Quick Charge, though it does take a pretty clear stand, “strongly recommend[ing]” against it. At issue, among other things here, is potential compatibility issues with standard USB chargers. Google wants everyone to play nicely together – that’s part of the appeal behind USB-C standardization in the first place. In fact, the document goes so far as explaining that it might make it a requirement in some future update. It minces words a bit less here, While this is called out as “STRONGLY RECOMMENDED”, in future Android versions we might REQUIRE all type-C devices to support full interoperability with standard type-C chargers. It’s essentially the technical paper version of “don’t make me put my foot down.”
|
Genesis Partners spins out $50 million fund, F2 Capital, to back early-stage startups in Israel
|
Lora Kolodny
| 2,016
| 11
| 30
|
Three members of the senior investing team at Genesis Partners, a major Israeli venture firm, are stepping out with a new, early-stage fund of their own called For the unfamiliar, portfolio companies have been acquired by the likes of Apple, IBM, Microsoft and Sapiens over the past decade, and have gone public on the Nasdaq exchange in the U.S. as well. Some that may be familiar to U.S. tech professionals include the 3D sensing company PrimeSense which Apple acquired for a reported $360 million, and a newer startup called Seebo that helps corporations operating outside of tech, especially toy companies, to create smart, internet-connected products. While Genesis Partners was a $600 million fund for growth-stage deals, F2 Capital, founded by Eddy Shalev, Jonathan “Jonny” Saacks and Barak Rabinowitz, is in the midst of closing a $50 million debut fund to back seed stage companies. Limited partners in F2 are mostly family funds based in Israel, the U.S. and Australia, Rabinowitz told TechCrunch. The newly spun out firm is taking over the management of an accelerator program that was also started by Genesis Partners called Founded in 2011, the Junction used to focus on connecting entrepreneurs with their first big enterprise customers. HP Inc, SAP, MunichRe, Enel, and Tesltra are among the accelerator’s strategic partners. Since 2011, the Junction’s program has evolved. It is generally modeled after the Israeli military selection process for elite units known as the “Gibbush,” according to a press statement from F2. Two hundred and fifty applicants pitch and only five companies enter the six month bootcamp program. F2 now provides $100,000 of funding on a convertible note to those top five companies. The firm maintains an option to invest more in their later rounds. Previous participants in the Junction have included: Appsflyer, Honeybook, Simplee, ClarityRay (acquired by Yahoo!), KitLocate (acquired by Yandex) and Moment.me (acquired by Wix). The newly admitted batch of Junction companies includes: Rabinowitz told TechCrunch that outside of its deals from The Junciton, F2 Capital will not invest in any business to consumer apps, content or biotech companies. Instead, the firm will focus on what VC’s are calling “frontier” technology today, or tech that’s beginning to transform existing industries or create new ones. Rabinowitz said, “People sometimes think of frontier tech as hobbyist drones and VR games, the cool fun consumer stuff. But we are thinking about the cross-section of big data, AI and connectivity across multiple segments.” More specifically, the investor said he’s looking for startups developing sophisticated insurance technology, technology that can improve the user experience and development of virtual reality and augmented reality products, drones that do more than take a picture, connected vehicles and cybersecurity.
|
Google enacts stricter penalties on sites that continuously spread malware
|
Sarah Perez
| 2,016
| 11
| 9
|
Google has been users dangerous and harmful websites for many years by warning web surfers when they accidentally click on links that could lead them to sites that spread malware or attempt to phish for your private information. But many sites figured out how to work around Google’s policies. Instead of cleaning up their act, they would instead temporarily make adjustments so their site would appear to be compliant when Google verified it, then return to harming users once Google’s warnings were removed. Now, Google says it’s putting a stop to this behavior by classifying these sites as “repeat offenders” and enforcing stricter penalties. In the past, when a site was in violation of Google’s , , , it would display a warning to web searchers until Google could verify that the site was no longer harmful. This verification could be triggered automatically, or it could be made at the request of the site’s webmaster, in an announcement. But that’s changing. Starting today, sites that switch between compliant and policy-violating behavior will be labeled as “repeat offenders,” and will not be able to request re-verification for 30 days. That means for an entire month after being flagged and labeled, these sites will continue to show warnings to web searchers. These warnings indicate that web users should avoid the site in question, and redirect you back to Google’s search results. That leads to dramatically decreased traffic for the sites in question. Only after the month is up, will the webmaster be able to request a review of their flagged site. The idea here is to close a gap in protection that existed with the prior policy. It will no longer be lucrative for these sites to only pretend to go complaint, as they’ll instead be continuously flagged for extended periods of time. Google says the policy won’t apply to those sites that are hacked, only those that spread malware and other harmful content intentionally.
|
Crunch Report | GoPro Cuts 15% of Workforce
|
Khaled "Tito" Hamze
| 2,016
| 11
| 30
|
Tito Hamze, John Mannes
Tito Hamze
Chris Gates
Chris Gates TechCrunch C/O Tito Hamze
410 Townsend street
Suite 100
San Francisco Ca. 94107
|
Fitbit is buying troubled smartwatch maker Pebble for around $40 million
|
Jon Russell
| 2,016
| 11
| 30
|
It looks consolidation is acoming to the wearables space with Fitbit set to acquire smartwatch maker and Pebble, according to . The site reports that the acquisition is a “small amount” but there’s no word on exactly how much, or indeed how little. Further, The Information said that the deal will see Pebble and its products closed down over time, with Fitbit acquiring its assets, which include intellectual property and software. “We don’t comment on rumors or speculation,” a spokesperson from Fitbit told TechCrunch. Pebble is yet to respond to our request for comment. A source close to the company told TechCrunch that watch maker Citizen was interested in purchasing Pebble for $740 million in 2015. This deal failed and before the launch of the Pebble 2 Intel made an offer for $70 million. The CEO, Eric Migicovsky refused both offers. Our source said that Fitbit is now paying between $34 and $40 million for the company and is “barely covering their debts.” Pebble released in October, but the past year or so has been a challenging period. , while was in some trouble and had turned to debt funding and loans, as well as traditional investor cash, “in order to stay afloat.” Earlier this year, Pebble CEO Migicovsky confirmed that his company had raised $28 million in debt and venture financing. He blamed a more cautious outlook from VCs focused on tech as the primary reason for letting 40 of Pebble’s staff go. Fitbit, too, has experienced its own challenges. when it listed on the New York Stock Exchange in 2015, but today it is trading at $8.40. That depression is largely down to less-than-impressive financial results. Some may cite the emergence of Apple and the Apple Watch as a competitor, but as initial consumer interest in wearable devices has waned. UPDATE – Pebble’s official Twitter account seems to confirm the news. https://twitter.com/Pebble/status/804180569848324096
|
Sensor-studded suit helps track recovery of stroke patients
|
Devin Coldewey
| 2,016
| 11
| 30
|
Millions of people suffer from strokes every year worldwide, and severe ones can require long-term care — but as with other rehabilitation regimens, it can be hard for doctors to track the everyday activities of their patients. A high-tech suit created by a Dutch grad student may be a powerful new way to help keep better tabs on those in recovery. Bart Klaassen, from the Netherlands’ University of Twente, and the rest of his team pursued it as part of a project resulting from Europe’s FP7 research initiative. Klaassen’s doctoral thesis is based on the work, as well. The idea is that while patients are closely monitored while at the rehabilitation clinic or in regular checkups, real-world tasks such as getting out of bed, navigating the house and cooking can only be reported secondhand. “There has long been a great need for systems like this, but the technology simply was not ready,” explained Klaassen . But as other projects along these lines have proved, that’s not the case any more: sensors can be made small enough, fabrics smart enough and the resulting data coherent enough that a high-tech suit like this is not only a possibility, but potentially a major advance. The suit, which is worn underneath the clothes for several months, contains 41 sensors that track strength, flexibility, gait and other relevant metrics. They’re collected wirelessly and processed by (at present) university servers. “We have succeeded in modelling all of the relevant movements, and in cleaning up the data that is relevant for the therapist by filtering out the rest,” said Klaassen. “Our project has delivered new techniques and methods that can be used to monitor patients at home for extended periods of time, and to identify any differences with structured clinical measurements.” Presumably the sensors and data could also be used for tracking progress in cases other than those suffering from strokes, but Klaassen’s research is focused on the one domain for now. If it proves useful it could be adapted for use in other forms of physical therapy. The suit isn’t in production or anything right now and there are plenty of questions, from whether it can be made cheaply to whether it can be washed at all — I’ve emailed the university with some questions and will update the post if I hear back.
|
Online real estate service OpenDoor raises $210M Series D despite risky financing model
|
John Mannes
| 2,016
| 11
| 30
|
This afternoon, announced that it led a massive $210 million Series D investment round in . With the new capital, the startup is looking to expand the usage of its marketplace platform for buying and selling real estate to 10 cities. OpenDoor is unique in that it owns its own inventory of homes. While the predictive analytics the company employs to project home resale value are complicated, the experience for buyers and sellers is fairly streamlined. If you opt to sell your property on OpenDoor, you will be given a valuation from the startup. Once you accept it, OpenDoor pays you for your home and effectively “flips” the real estate, seeking to sell it for a profit. To entice buyers, OpenDoor allows for self-guided property tours at any time, made possible by smart locks and security cameras. If you buy a house from OpenDoor, This time last year, the company raised an $80 million Series C. Across all rounds, OpenDoor has now raised a cumulative $320 million. It’s very likely that OpenDoor now holds a valuation that exceeds the $1 billion unicorn threshold. Other investors, including , , , , , , , and participated in today’s round. Khosla’s involvement is notable because OpenDoor executive chairman Keith Rabois is a partner at the venture capital firm. According to CrunchBase, Khosla Ventures has been invested since the company’s initial Series A back in 2014. Prior to this round, of OpenDoor, though it’s uncertain whether this remains the case after today. Beyond venture capital financing, Norwest disclosed that OpenDoor has “hundreds of millions” of dollars of debt to its name. This debt is what the company uses to actually purchase properties. In the banking world, there is generally an expectation that fintech companies and marketplaces prove out their business models before they can be issued debt. If this sounds too good to be true, you are in good company. has focused on the possibility of an economic downturn. Most bluntly, to companies that own large portfolios of unsold homes during a period of economic instability. [graphiq id=”laBaNqDFeQJ” title=”Lennar Corporation (LEN) Stock Price – Historical” width=”600″ height=”548″ url=”https://sw.graphiq.com/w/laBaNqDFeQJ” link=”http://listings.findthecompany.com/l/8029436/Lennar-Corporation-in-Miami-FL” link_text=”FindTheCompany | Graphiq” frozen=”true”] The startup believes that a “frictionless” marketplace reduces its overall risk. The argument follows that homeowners would be willing to sell their properties for whatever they could get in a complete meltdown and OpenDoor would get a cut no matter what — assuming the team’s data analytics group is up to the nearly impossible task of predicting the future. To date, the company has 200 employees servicing both the Dallas Fort-Worth and Phoenix markets. These regions alone account for approximately $60 million in transaction volume per month for OpenDoor.
|
Tesla will move to open Richmond store after DMV win
|
Darrell Etherington
| 2,016
| 11
| 30
|
Tesla has won the right to open and operate its own dealership in Virginia, after the state Department of Motor Vehicles commissioner reversed a ruling from September denying it a license. In his decision, the commissioner cited overwhelming public support, as well as unique sales and service requirements as reasons for his finding that “no dealer independent of Tesla” in proposed store city Richmond is fit to do so correctly. Tesla’s been fighting for the right to open and run dealerships for its vehicles across the U.S., in what has become a state-by-state battle fought in local courts and regulatory bodies. The Tesla win in Virginia is significant, but it isn’t a total green light; the carmaker still needs approval from the state’s Motor Vehicle Dealer Board — still, all indicators suggest it’s going to be able to open and operate the Richmond location soon. While the state’s Automobile Dealers Association lobbied to prevent the license being granted, the commissioner noted in his ruling that even by their own admission they would not be able to make money on operating third-party Tesla dealerships given the carmaker’s sale model. The commissioner also cited Tesla’s leading customer satisfaction stores for the quality of its service centers as another reason for granting the license. Another reason given for allowing the license is the distance prospective Virginia Tesla customers would have to travel to find a store, and then to service one post-purchase. The full text of the decision is embedded below. A Tesla spokesperson provided the following statement to TechCrunch when contacted: Tesla applauds the Commissioner’s decision to allow us to open our new store and service center in Richmond, Virginia. This decision will allow Richmond-area consumers to learn about and purchase their Tesla vehicles in closer proximity to their homes. We are thankful for the many Tesla owners and Virginians for their support. We intend to swiftly begin construction to open our new store and service center at 9850 West Broad Street and look forward to joining Richmond’s business community.
|
Counting endangered sea cows is hard, so we’re going to make AI do it
|
Devin Coldewey
| 2,016
| 11
| 9
|
Can you spot the lone dugong in the image above? Look closer. Closer… still no? Try this . Give up? Now do that with 45,000 more, and you’ll have a general idea of the population of these endangered critters. If that sounds tedious, then perhaps you, like researchers at Murdoch University, would prefer to . , of the school’s Cetacean Research Unit, has been using UAVs to capture images of marine animals for years, but the data piles up fast, and there are only so many grad students. This one’s a little easier. Hodgson worked with computer scientist Frederic Maire, of the Queensland University of Technology, to automate the process. They trained a system on imagery where the sea cows had already been tagged, allowing it to look at fresh photos and spot them with about 80 percent accuracy. The success rate should improve, and at any rate it’s good enough to give a ballpark measure and indicate images that need to be scrutinized more closely. The one below, for instance, may need to be double-checked, since there are so many creatures in it and the shadows may fool the system into counting a few twice: With some tweaks, the system could also be convinced to watch for whales, dolphins, boats, and other common features of the coast that could stand to be quantified. Just one more way that machine learning and computer vision are helping out scientists and others across the globe.
|
Motorola ‘likely’ to launch a Project Tango module for the Moto Z
|
Brian Heater
| 2,016
| 11
| 30
|
Slowly but surely the Moto Z is getting an ecosystem. The promising handset launched with some solid specs and a handful of add-ons that made it the most fully formed modular handset to date — which, granted, isn’t saying a whole heck of a lot. And Motorola’s been working to add to those offerings by opening up its Mods to developers and even launching a handful of hackathon designs to bring functionality to the back of the handset. At an event in Chicago this week, the company’s president Aymar de Lencquesaing that the Z handsets are “likely to have a Tango module.” A little wishy-washy perhaps — and certainly short of a full-on announcement — but it’s a partnership that makes sense. After all, Motorola’s parent, Lenovo was the first to really embrace Google’s augmented reality camera technology in the form of its Phab 2 Pro [pictured]. And while Tango in its current form isn’t a strong enough platform to be the deciding factor in the purchase of a handset, it could work well as another add-on for the versatile phone. Housing it as a module would also go a ways to addressing the additional hardware (additional cameras, more battery) to really power the indoor 3D mapping software. It could also help secure the Moto Z’s place as a platform for hardware experimentation in ways that just aren’t possible on the current two-year smartphone upgrade cycle.
|
Skillz wins two new patents, is now helping brands sponsor e-sports tournaments
|
Lora Kolodny
| 2,016
| 11
| 30
|
Game tech startup today launched a service that allows brands to easily sponsor and host mobile e-sports tournaments for any game titles that they like, as long as they are already part of the Skillz platform. As we’ve previously reported, Skillz enables developers to turn mobile games into tournament-playable titles without writing a ton of new code. Its technology operates as a kind of digital referee, making sure competitors of like skill are matched in competition, and that players face equal challenges. Skillz also enables games to be played in compliance with all local and federal regulations for cash, prizes or points. The company operates something of an online catalog of tournament-playable games, in which brand managers could easily discover titles they may like for potential sponsorships. Skillz executives say that making mobile games into competitive ones will help developers generate more revenue, but also generate excitement and longer-term engagement in use of their entertaining products. The company will split sponsorship revenues between brands and games on its platform. Specific terms of each sponsorship will depend on the popularity of a given game. Skillz CEO and founder Andrew Paradise Besides launching their brand sponsorships feature today, Skillz was recently granted two new patents, bringing the company’s patent wins to four total. Skillz CEO and founder Andrew Paradise said the patents cover two core elements of his company’s technology. “One is for the algorithm that enables Skillz to evaluate how much an outcome in a game is based on skill versus chance. It is based on research at Dartmouth and solves a big problem for the e-sports industry,” he explained. There has been significant debate and scrutiny paid by regulators around what constitutes a game of chance versus a game of skill online. Cash prizes are available for tournaments where games of skill are involved, but cash prizes for games of chance would be defined as gambling online, which is not legal throughout the U.S. today. “The second patent we won,” Paradise said, “has to do with our integration technology. We have a allowing them to integrate, self-service and implement e-sports into their games. We call it a tournament management system, or TMS, which is analogous to a content management system that you’d use to run a company website or other publication online.” Skillz news comes as publicly traded companies like Activision, Electronic Arts, Amazon and even Facebook and Google are all making a push into e-sports. More than 3,000 game studios are using Skillz’s platform to enable e-sports, or tournament plays, of their games today, the company reports.
|
null |
Sarah Perez
| 2,016
| 11
| 9
| null |
Losing the future
|
Anthony Ha
| 2,016
| 11
| 9
|
There’s a story that Silicon Valley likes to tell about the future. Part of that story, of course, is the idea that technological change will continue to accelerate and make our lives better. But when I talk to entrepreneurs and investors, I hear a broader narrative of progress — a belief that society is becoming more diverse, meritocratic and rational. Things are , and we, as a community, are . That’s the assumption behind so many conference pitches, where someone talks about how something hasn’t changed in umpty ump years and isn’t that ? “We’re going to disrupt Industry X,” we’re told again and again and again (yes, ), and we’re meant to assume that this disruption will be a good thing. Heck, when people ask me why I write about technology, I usually tell them it’s because I want to write about the big forces of change. I also hear echoes of that narrative when I discuss the challenges facing diversity in the tech industry, and then friends and acquaintances assure me that (the industry’s young-ish, educated, urban professionals) don’t think that way. Soon, they say, will be like us. That’s the narrative that Donald Trump just punched in the face. To be clear, I’m conflating a few different things here — economic growth, technological progress and political change. But this combination is part of what we’re talking about when we point to . And the ties seem particularly clear in this election, where Hillary Clinton attracted . (Well, .) Where she reportedly ran a . (And yet …) Where she, , has been frequently .” And where, most importantly, the candidates offered two diametrically opposed visions of the future. I expect that we’ll be reading post-mortems about why Trump won for months and years to come: To what extent was it about economics? Or race? Or gender? Or all of the above, in a white, working class uprising? But here’s what seems clear to me: Trump, for all his , told a consistent, resonant story. He told Americans that we are at “ ,” that the changes that many welcome as progress actually reflect a once-great country in decline. That people are right to fear immigration, to fear growing protections for minorities and the LGBT community, to fear abortion rights, to fear — the disruption brought by free trade, by automation, by new technologies. He told us that things aren’t getting better. And it seems that more than 59 million people were ready to hear that story. What did Clinton and the Democrats offer in response to that fear? “America is already great.” They told us that we’re making progress, and that the problems that remain can be fixed with the right combination of policy, data — . Don’t get me wrong: The case for progress has its own believers, enough to . And despite plenty of my own reservations, it’s the story that I, fundamentally, believe. Or at least I believed it yesterday. I hope I’ll be believe it again sometime soon. But right now, it rings a little hollow.
|
Referees will wear hats with built-in GoPros during the Big Ten Championship Game
|
Fitz Tepper
| 2,016
| 11
| 30
|
Are you upset your team isn’t playing in a championship game this weekend and need a reason to watch? Well here’s a reason to watch at least one game on Saturday: Referees will wear hats with built-in GoPros during this weekend’s NCAA Football between Penn State and Wisconsin. The project is part of a deal between FOX Sports Lab and GoPro to offer new viewing experiences to viewers. It’s not yet clear if the hats will just have a store-bought GoPro attached, or be a custom-built wearable with an integrated camera and wireless system. Footage captured from the camera-hats will be shown on FOX during the broadcast, as well as made available online and on FOX Sports’ social media accounts. Check out the clip below to see footage from a trial run during the Wisconsin-Minnesota Big Ten game last weekend: It’s pretty amazing. The hat-cameras are able to provide a close up view that we’ve never seen before during football games — it almost makes you feel like you’re playing a game of Madden on Xbox. For now this seems like a one-time thing, but if you watched the footage above it’s clear this could be the future of watching professional sports. These cameras could also help catch missed calls during challenges and replays in football, which is currently a hot topic in the NFL. This isn’t the first time GoPro has worked with major sports leagues. GoPro also to capture footage during games and practices, but that footage was shown after the game — not live on a network. The Big Ten Championship Game will be on FOX at 8:17pm ET on Saturday.
|
Oldest surviving Nikon fetches $406K at auction
|
Devin Coldewey
| 2,016
| 11
| 30
|
A fabulous piece of photographic history just sold at auction for $406,000, more than twice the estimated haul. This Nikon One was the third production camera — as in the third it built — from the now-famous company, and it’s the oldest one known to still be around today. Nikon began as a optics company, Nippon Kogaku K.K., in 1917, and made a few lenses and even worked on a Canon body in the 1930s. But 1948 was the first year it made its own cameras, starting with the One or I, of which less than a thousand were made. (A bit more history .) It’s an interchangeable-lens rangefinder, and the lens included is also an old one: the 11th Nikon ever made, a 50mm f/2 that collapses back into the body. It comes with a great old double-strap leather case that I would definitely like a replica of for my Olympus. While it’s certainly an interesting piece of history, it probably won’t be doing much shooting: It used a non-standard film size that limited its appeal, something Nikon fixed in following models. [gallery ids="1422394,1422395,1422396,1422397,1422398,1422393"] Engraved on the bottom, rather prominently, is “MADE IN OCCUPIED JAPAN,” a rather chilling reminder of the times in which this camera was made. The starting price for this historic camera was €90,000, and it was estimated to go as high as €180,000; it eventually sold for €384,000, (approximately the aforementioned $406,000), inclusive of the fees and premiums. You can . DPReview , including a Leica camera-rifle and a wonderful-looking M3 owned by Herbert List.
|
Apple acknowledges the new Calendar spam problem, says it’s working on it
|
Greg Kumparak
| 2,016
| 11
| 30
|
If you use iCal and have noticed a weird, sudden influx of calendar invites suggesting you buy bootleg purse knockoffs… . Just before Black Friday, spammers started tapping a relatively new technique: sending spam as event invites to iCloud email addresses, rather than sending standard emails. By default, iCloud processes these and tries to add them to your calendar… which helped the spammer sneak around spam filters generally caused a notification to pop up on your various screens. It’s not a particularly malicious trick, but damn is it annoying. After a few days of silence on the matter, Apple has acknowledged the issue, apologized for the spam and says it’s working on fixing it. Here’s the statement from Apple: We are sorry that some of our users are receiving spam calendar invitations. We are actively working to address this issue by identifying and blocking suspicious senders and spam in the invites being sent.
|
Founder Collective easily finds $75 million for fund three
|
Connie Loizos
| 2,016
| 11
| 30
|
, the seven-year-old, Cambridge, Ma.-based seed-stage venture fund, has closed its third fund with $75 million in capital from a small group of limited partners whose anchor investors are the the firm’s three partners: David Frankel, Micah Rosenbloom and Eric Paley. “Unlike most funds, the partners are, in fact, the single-largest investor in [our new fund]. So when entrepreneurs take our money, it’s really money,” says Rosenbloom, who earlier in his career co-founded a 3D imaging startup that to 3M. Given Founder Collective’s success to date, we’re guessing that’s very much by choice. Consider that it backed the adtech company The Trade Desk, which managed to in September. It was also an early investor in Cruise Automation, to GM earlier this year for a billion dollars; the 3D printing company MakerBot, by Stratsys in 2013; the live streaming startup Periscope, by Twitter last year; the backend service Firebase, by Google in 2014; and the private cloud service Blue Box, to IBM last year. Did we mention the outfit — which closed its with roughly the same amount — has stakes in privately held Uber, along with BuzzFeed, HotelTonight, ThredUp, FormLabs, Optimizely, PillPack and SeatGeek? Earlier today, we chatted with the three partners about their new fund, and why they didn’t opt to invest a bigger fund this time around. DF: Ninety percent of our investments are in four geographies: San Francisco, New York, Boston and Southern California. That said, we’ve invested in companies in South Africa, the U.K. and Israel. One of our best investments ever is a company called , South Korea’s answer to Amazon, which recently raised . MR: That founder-investor alignment really matters, meaning fund size, follow-on strategy and just the chemistry match of founder and investor. Many founders don’t understand their investor’s strategy as well as they should, but it will inevitably impact their company. We try to be really explicit about [our positions on things] from the get-go. MR: We’ve learned that party round syndicates with no clear lead are bad for entrepreneurs and for us as investors. When no investor owns the outcome and has responsibility for helping the company, often the founders get no investor support from anyone. We’ve found ourselves stepping into that vacuum, but the lack of investor advocacy underscores a long-term problem for the company. EP: We’re pretty opinionated about this and were thrilled to write a couple of guest posts for TechCrunch on it. We think based on data from . If you look at the last five years of IPOs, the companies that raised the least money outperformed the companies that raised the most on the public market. Our data showed there was no correlation between the amount of funding a startup received and its ultimate performance. Capital doesn’t come with insights. If you can’t turn $1 into $10, it’s hard to turn $1 million into $10 million. Overfunding hurts startups even in the best outcomes, and we don’t think it’s great for VCs, either. EP: We typically write a single follow-on check in the Series A round and that’s it. We’re not life cycle investors and think that focusing on later rounds actually dollar-cost averages your returns down while misaligning investor interests with founders. We really don’t focus on ownership percentage. We think it’s the tail wagging the dog. DF: We think some of the best investors are busy running companies full time. We also think that often the best entrepreneurs tend to check in with aspirational founders before soliciting capital. That’s why we’ve always tried to work closely with founders who are really well-known and respected in their communities. On a practical level, our founder partners invite one of us into a pitch meeting and actively engage in the decision-making process with us. We so highly value conviction internally, that if one of them is banging the table, chances are high that we’ll make the investment together. MR: We’ve opened an SF office, but everything else is the same as the past two funds. DF: They’ve risen a bit, but looking at our data, we’re surprised by how little. We’re not sure if that reflects the macro environment, or just our aversion to deals with big sticker prices. EP: I’d say 2009 to 2010 was a pretty dark time, as well, and that’s when we had the good fortune to invest in Uber and The Trade Desk, so we’re big believers in the ingenuity of entrepreneurs no matter the macro conditions. In fact, tech might be one of the most important ways we can preserve and enhance our freedoms.
|
MusicNet aims to give machine learning algorithms a taste for Beethoven
|
Devin Coldewey
| 2,016
| 11
| 30
|
We’ve got machine learning systems captioning images, synthesizing speech and translating between languages, but it seems we’ve been neglecting their cultural development. Would you believe the future rulers of this planet are largely ignorant of classical music altogether? It’s bad enough humanity will be enslaved — but by an entity so vulgar? A new project looks to change that with a data set of pieces from great composers, annotated so the computers can follow along. , is an attempt to create a data set that is to analyzing classical music what the widely used ImageNet is for computer vision. It isn’t the first time machine learning has encountered music, by a long shot, but MusicNet is more about establishing a standard set of data to learn from and score by. ImageNet, for instance, is used by researchers and students both to train and evaluate computer vision systems, and other data sets can be used for translations, faces and so on. To start with, MusicNet comprises 330 classical music recordings available for free online from a number of sources; it’s heavily weighted toward Beethoven and solo piano because, well, Beethoven’s piano work is pretty popular. But there are also dozens of works by Schubert, Brahms, Mozart, Bach and so on — though, disappointingly, no Chopin. Each live recording is annotated by matching it to a score, with each note indicated down to the millisecond. That would normally be an extremely labor-intensive process, but the team used a technique called dynamic time warping to map the ideal written score to each performance, in which creative liberties and the innate variability of human execution cause digressions from the original. The automated approach (still manually checked by human musicians) both improves precision and makes processing music far less laborious. “Features are traditionally hand-crafted in the the music community,” wrote John Thickstun, lead author of , in an email to TechCrunch. “This is analogous to that state of affairs in the vision community a decade ago, before deep models trained on datasets like ImageNet replaced visual hand-crafted features with learned ones.” The team set a few machine learning systems to the task as a test run: the models were trained up on the MusicNet data, which would presumably allow them to associate notes with aspects of the recording associated with them. Visual representation of how the model extracts basic note data from raw audio (whole note on left, weighted center in the middle, analysis of that weighted area on the right). “We asked our networks to guess which notes are played in segments of recordings, and the networks discovered features they find useful for this task,” explained Thickstun. Early results are heartening: Models performed better when trained on MusicNet than on traditionally annotated pieces, producing more accurate note maps when presented with novel recordings. Just as computer vision systems began only being able to say whether two images were the same or different, or whether an object was square or not, the nascent music analysis models are in their infancy. Basic tonal differentiation is a start, but of course there’s much more to a piece of music than “B for 150 milliseconds, C+E for 300 milliseconds,” etc. “In future work, we plan to build deeper networks that learn more complex features,” said Thickstun. “We hope these networks will be able to learn higher-level musical concepts like melody, harmony, and rhythm.” Adding more music to the data set, which the creators urge other researchers to do, will help broaden and deepen the understanding of any model trained on it. Eventually, as other neural networks and deep learning systems , they may one day create their own works. You can read more about the methods and models behind MusicNet in , or at .
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.