Don Coolidge and JP Benini are bringing supercomputer smarts to the world of children’s toys.
Coolidge and Benini just launched a Kickstarter for a toy dinosaur toy driven by IBM Watson, the machine learning service based on the company’s Jeopardy-playing supercomputer.
Developed under the aegis of a company called Elemental Path and a project called CogniToys, this tiny plastic dinosaur uses speech recognition techniques to carry on conversations with kids, and according to Coolidge and Benini, it even develops a kind of smart personality based on likes and dislikes listed by each child.
The toy is another example of online machine learning pushing even further into our everyday lives. This is made possible not only by an improvement in AI techniques, but also by the ability to readily deliver these techniques across the net. The CogniToys dinosaur connects to a Watson cloud computing service via the internet. “If we had relied on doing all this with the hardware,” says Coolidge, “it would have become a really expensive toy. It would be unaffordable.”
Parents connect the toy to a home Wi-Fi network, and then they input some details about their child, including such things as age, grade level, favorite color, sport, or food. This helps the toy interact with the child, but using Watson, it can also evaluate a child’s ability and skill level on its own.
“If your kid is, say, using new vocabulary words, we can bump up the skill level on the assessments to push the child further,” Coolidge says. A child in the first grade, he says, would use the toy differently from someone in the second grade.
For Coolidge, who started developing the toy as part of Watson-centic contest sponsored by IBM, the added benefit with this internet-powered toy is that his company can continue to improve the toy after it’s in the hands of kids. “Just like app developers,” he says, “we’ll be able to respond to user feedback and make the product better.”
Microsoft continues to evolve into a company that plays nicely with competitors.
Today, the company announced that its Office apps for Apple iPhones and iPads now provides tools for opening and editing files stored on third-party online services from Apple and Box.com. And it unveiled a new Cloud Storage Partner Program that will let third party apps to open and edit Office files via its internet-based Office Online service.
These are small changes, but they’re part of an enormous shift for Microsoft, which has a history of working to ensure that Microsoft tools only worked with other Microsoft tools. Following the appointment of CEO Satya Nadella, the company has not only dovetailed its software third-party services, but also released software that runs on rivals operating systems, including Apple’s iOS.
Back in November, the company unveiled a partnership that lets Office apps directly save and sync files with Dropbox, and now it’s expanding this same kind of thing to other services. Meanwhile, Box, Citrix, and Salesforce are working to build apps that can edit files via Office Online.
A new study suggests that bacteria may be able to jump between host species far easier than was previously thought. Researchers discovered that a single genetic mutation in a strain of bacteria infectious to humans enables it jump species to also become infectious to rabbits. The discovery has major implications for how we assess the risk of bacterial diseases that can pass between humans and animals. It is well known that relatively few mutations are required to support the transmission of viruses -- such as influenza -- from one species to another. Until now it was thought that the process was likely to be far more complicated for bacteria.
Scientists at the universities of CEU Cardenal Herrera (Spain) and Glasgow and Edinburgh (UK) studied a strain of bacteria called Staphylococcus aureus ST121, which is responsible for widespread epidemics of disease in the global rabbit farming industry. The team looked at the genetic make-up of ST121 to work out where the strain originated and the changes that occurred that enabled it to infect rabbits. They found that ST121 most likely evolved through a host jump from humans to rabbits around 40 years ago with a genetic mutation at a single site in the bacterial DNA code the cause for this.
The discovery transforms our understanding of the minimal genetic changes that are required for bacteria to infect different species. ST121 is found in the respiratory tract and on the skin of some people. While it is usually harmless, the bacteria can cause a variety of conditions from minor skin infections to meningitis and sepsis. In rabbits, the bacteria can cause serious skin infections.
Professors David Viana, of CEU-UCH Veterinary Faculty, and Jose Penades, of theInstitute of Infection, Immunity and Inflammation at the University of Glasgow, who co-led the study, said: "The ability for pathogens to switch host-species and lead to an epidemic in a new host population is of major concern to veterinary and public health professionals. Our results represent a paradigm shift in understanding of the minimal adaptions required for a bacterium to overcome species barriers and establish in new host populations."
Professor Ross Fitzgerald, from the University of Edinburgh's Roslin Institute, who co-led the study, said: "Domestication of animals, industrialisation of agriculture and globalisation have provided new opportunities for the transmission of bacteria between humans and animals. This latest research has important public and veterinary health implications which will require a re-examination of the future threat posed by bacterial host switching events."
Story Source:
The above story is based on materials provided by Asociación RUVID. Note: Materials may be edited for content and length.
Inside the San Francisco offices of Pivotal, a business software outfit spun off from big-name tech companies VMware and EMC. Photo: Ariel Zambelich/WIRED
Pivotal—the rather ambitious business software outfit spun off from big-name tech companies EMC and VMware—is open sourcing three of its key products, sharing the underlying software code with the world at large.
Today, the San Francisco-based company announced that in the coming year, it will open source GemFire, HAWQ, and GreenplumDB, three “big data” tools designed to help businesses analyze large amounts of digital information. “Our customers are starting to look to open source, and they’re looking to projects that have communities around them,” says Pivotal’s Sunny Madra, who oversees the company’s data work. “Customers want a say in the direction of software.”
The move is yet another sign that the world of business software is changing, moving away from proprietary software built and licensed by individual vendors like Oracle and Microsoft, towards open source tools that anyone can freely use and modify. With so many businesses now erecting their online operations atop open source software, we’re even seeing the traditional vendors change their approach.
Microsoft is open sourcing some of its key software tools. And Pivotal, with its roots in old-school software companies VMware and Greenplum, is another notable example. The company previously offered an open source cloud computing tool called Cloud Foundry as well as software based on the open source data-crunching software Hadoop, and now, it’s open sourcing the rest of its major tools.
“Pivotal is doing this in a way that a broader foundation of vendors and end using can collaborate on this software,” says Shaun Connolly, vice president of corporate strategy at open source software company Hortonworks, which plans to work in tandem with Pivotal on these tools. “The expectation this days is that software be driven through an open source model.”
Pivotal plans to open source HAWQ and GreenplumDB through the Apache Software Foundation and push the GemFire code into the community that oversees the open source database PostgreSQL. This should mean not only that the code is freely available to anyone, but that a broad community of developers—some inside Pivotal, others outside—will drive the evolution of these tools. Like many other open source software companies, Pivotal will make its money by helping businesses use the software and by selling specialized versions of the tools.
According to Madra, Pivotal’s big-data tools pulled in over $100 million in bookings from business in 2014. But he also says that businesses increasingly want to spend money on software that isn’t controlled by a single vendor, and he says that Pivotal has already built successful businesses around open source tools such as Cloud Foundry. “Open source,” Madra says, “is a trend growing bigger by the day.”
Recently, a computer science student at the University of Illinois did some class homework and posted the answers to GitHub, the code-sharing platform widely used by open-source software developers. And the University was peeved.
Last week, using a DMCA takedown notice, the standard way to request removal of copyrighted material from the net, the University tried to force GitHub into vanishing the coursework from its service. After criticism from students, the school has rescinded the notice, but the incident goes a long way towards describing how the software world has changed in recent years.
In short, the world’s developers are moving towards a model of open collaboration. And though that works well for them, it clashes with the way the world of programming traditionally operated—as embodied by the University of Illinois.
A Noble Enough Idea
The University wanted to remove GitHub repositories that contained “course materials and solutions to assignments in three of our CS courses,” according to Professor Rob Rutenbar, the head of the school’s computer science department. The implication is that school was trying to stop cheating.
That’s a noble enough idea. But in this case, it works against the prevailing attitudes in the software world.
Across Silicon Valley and beyond, GitHub has caught on big time because it lets you track changes to your work—a version control system, in computer parlance—and it lets you do this in the open. This is a great way to promote collaboration on software projects, and naturally, the general attitudes behind the site are now filtering down the country’s university students.
The trouble is that while students often think of programming as a collaborative act, some universities still think of it as a private, proprietary thing. “The whole concept of using a DMCA takedown on the student who is openly working is broken,” says Andrew Dunn, a graduate student at Northwestern University.
Learning to Merge
Yes, the clash is a little more complicated than that. “They’re stuck with a damned-if-you-do, damned-if-you-don’t situation,” Mike Berry, a computer science professor at the University of Tennessee, Knoxville, says of his counterparts at the University of Illinois. “They don’t want to suppress, but at the same time its an academic integrity issue.”
But clearly, this is about more than cheating. Many students publicly complained about the situation, and in the end, Rutenbar listened.
“Once we dug into it as an organization with our students, we recognized that the DMCA take-down requests weren’t the best way for us to approach this issue,” he says. “[W]e choose our students because they’re extraordinarily bright, thoughtful, and unafraid of speaking up. And they’ve spoken up on this one.” And rightly so.
It’s important to strike a balance between teaching open collaboration, and teaching kids how to code, Rutenbar says. “We’re taking very young students who often come to us with some basic software skills, developed in a collaborative environment, and experienced in use of the internet as the library to find useful bits and pieces” he says. “Then we put them in a class where we need to carefully assess each student’s individual understanding of the material”
Better Borrowers
Twenty five years ago, when Mike Berry was a student, the idea of collaborative coding was virtually unknown. “Years ago we had nothing. So we learned by creating,” he says. “Now you don’t have to learn by creating. You can learn by merge. In some sense it becomes a little bit more of an engineering exercise.”
What that means is that today’s students need to be better borrower. Finding tools out in the open and adapting them to fit their particular problems. That’s made today’s programmers polymaths of sorts, able to tackle problems in new fields—big data, artificial intelligence, bioinformatics—that would have required years of specialization a few decades ago.
But they also have to be able to do the programming, says Berry. “The question mark is do you have the critical analysis,” he says. “Do you have the ability to discern whether the code is doing what it needs to do for your purposes; and how do you do that unless you know how to write it yourself.”
Berry and Rutenbar are now experimenting with testing students on computers that are completely disconnected from the internet. They use them to find out if their students can code on their own. And you won’t pass if you cheat your way through college using GitHub.
There are two devices: A smart Hub that resides on the wrist, and Pocket, a screen that looks like a smart phone. Neptune
There are two devices: A smart Hub that resides on the wrist, and Pocket, a screen that looks like a smart phone.
Neptune
But the Pocket is not meant to be a smart phone. Rather, it's simply a screen that the Hub can stream to, when a bigger screen is desired. Neptune
But the Pocket is not meant to be a smart phone. Rather, it's simply a screen that the Hub can stream to, when a bigger screen is desired.
Neptune
The Hub comes with 4G, wi-fi, NFC, GPS, and will run on Android Lollipop. Neptune
The Hub comes with 4G, wi-fi, NFC, GPS, and will run on Android Lollipop.
Neptune
Part of the reason the Hub can pack so much capability onto the wrist is that the electronics along its entire length. Neptune
Part of the reason the Hub can pack so much capability onto the wrist is that the electronics along its entire length.
Neptune
So far, according to the designers and engineers, Neptune's concept is far past the proof of concept stage. The circuitry has been laid out and the parts have all been identified. Neptune
So far, according to the designers and engineers, Neptune's concept is far past the proof of concept stage. The circuitry has been laid out and the parts have all been identified.
Neptune
Today, with every other commercial pushing a new phone or tablet, it’s easy to forget how drastically things can change, and how quickly. The iPhone, after all, is just eight years old—and five years ago, almost no one was talking about how mobile computing was going to eat the world. And yet today, the smartphone is the center of our digital lives. Does it have to be?
Neptune, a new startup launching today, believes the era of the smartphone is headed towards an end. “The current wearables space is best summarized as a master/slave relationship,” says Simon Tian, Neptune’s boyish founder, who is all of 20 years old. In his formulation, the phone is the master, and all wearables are the slaves, simply borrowing from the power and capabilities of the phone. “The smart watch actually just proves you’re also a slave to your phone. Imagine when it’s out of sight. You freak out!” Tian says.
Neptune wants to switch the relationship. Its wrist-worn device, the Neptune Hub, runs on Android Lollipop, and packs in the capabilities and power of a smart phone, with a quad-core processor, 4G, Bluetooth, wi-fi, GPS, and NFC. It’s meant to be the center of its wearers digital life. Instead of a phone, the Hub comes paired with the Neptune Pocket, a 5″ screen that’s available anytime a more robust display and touchscreen is called for, and has an 8mp camera on the back and a 2mp camera on the front. The Pocket isn’t a phone as much as it is a screen to which the hub can stream its capabilities. (When the Hub is low on power, the Pocket also serves as a juice pack.)
The move of making the wrist-work device smart and making the pocket screen dumb might not seem like a major switch, but Tian is quick to argue that many pain-points of our digital lives simply disappear, once the slave breaks its chains. The security of your phone itself is gone, since the Hub is always on your wrist. As for the pocket screens, those are low cost, easily replaced, and sharable between users. You simply pair them with your Hub, as needed.
Moreover, the Hub is meant to serve as a new “bare minimum”: The screen is big enough that you can glance at notifications and respond to messages; it can also track fitness and place calls. Thus, rather than having a phone that always seems to suck you in, you can simply do the main things you still need a phone for, in a far less burdensome way.
Origin Story
Tian, young as he is, has been dreaming of the Hub and Pocket since November 2012. The idea first in a rush, when he was still a teenager and fantasizing about starting his own business. The benefits just seemed too obvious. If you used a hub to stream content to a pocket screen, you wouldn’t have to worry about pesky problems like continuity between devices. Hell, the pocket screen didn’t have to be particularly powerful at all.
But the technology didn’t actually exist. So instead, as an exercise, Tian resolved to ship a smart watch, just to learn the ropes of hardware production. He called it Pine, and, in January 2013, threw some half-baked renderings up on a website that he registered for. He admits that, at the time, he had no real sense of what it would take to make them real. People mocked the idea: “Apparently this runs on magic,” said one tech writer. Yet the orders came flooding in. Tian dropped out of his pre-college program, and threw himself into the arduous task of scouring Shenzhen’s factory ecosystem, finding suppliers and partners. When those were in place, he raised nearly $1 million on Kickstarter. Against all odds, the Pine shipped out to over 7,000 customers in June 2014. It wasn’t a breathtaking piece of hardware. In fact, it was mostly a smart phone strapped to your wrist. But it was the world’s first stand-alone smart watch, requiring no cellphone tethering to function. And to Tian, it was tantalizing proof that with a novel-enough idea, he could boot strap a bonafide hardware company.
This time around, Tian not only has the core technologies to make the hub concept real, he has a bonafide team around him, anchored by Pearl, the design firm perhaps best known for developing the Misfit Shine. To date, according to Pearl’s founder, Mladen Barbaric, they have already moved well past the point of knowing whether the product is feasible. They have already finishing selecting components and engineering how they’ll fit together. They have already begun the arduous radio-frequency testing process. They already have have a slew of features and additions they plan on revealing in the coming months—including accessories, features, and how you interact with Hub.
For for now, they just need to see if there’s a market for the product. Tian believes their must be. “It has become so cheap to push a message that now it’s all about your idea,” says Tian. “The gap between a startup like ours and Samsung just isn’t as big as it was.” Neptune is wagering that in between the looming shadows of Apple, Samsung, LG, and Google, there’s just enough daylight for startup to sprout.
What is the goal of a university? Why do humans get college degrees? Here are some common options for why a human should go to college and earn a degree.
Humans should earn a degree so that they can get good grades. You might think this is a joke of a reason, but it’s not. Many people actually think this way.
Get a degree to get a job.
Get a degree to prepare you for a job (this is slightly different than the previous).
Go to college to learn to think – critical thinking skills.
Graduate from college with a degree to boost the university’s graduation rate.
College is all about the party and fun times. Meet some people too.
But what do I think is the reason to go to college? I pretty much agree with Chris Lee in this very insightful post about higher education. We like to say that the structure of courses in a university degree program are designed to promote critical thinking, but this isn’t always true. Historically, spending 4 years in college was a way for young adults to gain maturity before getting down to actual work (but not ditch digging work – you know, bank work and stuff).
Does this mean that it is worthless to go to college? Well, if your tuition is very high it might not be a good value. However, I think that people can still gain quite a bit by going to college and spending some time working on some difficult problems.
I always tell students: college is what you make of it, it’s not what college does to you. If you just sit back and take courses, then courses will be all that you have. If you participate in extra activities, and go to class, and do some research, and make some friends – in the end you will have something valuable.
Producing a Workforce
This seems to be a common theme these days. Administrators and politicians are emphasizing the importance of colleges in their role of producing an educated workforce. More people in college means more people that can work at higher level jobs. Clearly, I don’t think this is true (see above). Yes, humans with a college degree would be better at some jobs – but you can’t make a college degree job training.
But what if we did? What if we optimized the whole higher education system to train and develop humans to be work-ready? Here are some things that we should do:
Stop having college majors. Why major in math, physics or english? Those aren’t jobs, they are fields of study. Of course there are already some majors that have acceptable titles – accounting, nursing, and even biology-pre med. The courses in the majors would still have to change. Maybe business majors would be ok – not sure.
No more silly required courses. Why should everyone have to take algebra and a science course? If a student wants to be a high school english teacher, why should that student study algebra? Oh, they have to average grades and stuff? No, that is not true in the modern world. A computer could do it. Why should a marketing major take science? If they need any science ideas, they can just “google it”.
In fact, why have classes at all? If it’s job training, then there should be training courses. How do you use particular software packages that your career uses? How do you put together a TPS report? Things like this don’t need to be a semester long class, they could be shorter presentations.
No more grades. Instead of grades, there should be proficiency exams. If you pass, you move to the next skill-based test. Actually, this really isn’t a bad idea.
What about internships? Maybe the best way for students to be work-force ready is to just send them out to a job. Really, this would be win-win. The students would pay the university tutiion so that they could become an intern. The employers would get free workers and train them in the processes. Everyone gets something (really, this is win-win-win).
The Problems With Workforce Development
Perhaps I shouldn’t write a post like this. There is a small chance an adminstrator or politician will read this and say “Hey! What an awesome idea!”
I obviously don’t think this job training version of a college degree is a good idea. Here’s why:
A job training strategy only works for jobs that you know about. How do you train for jobs that don’t yet exist?
If you come up with a way to train the workforce, wouldn’t this just be pseudo-code for a robot? Once you really figure out how to do a job, a robot could do it for cheaper. This would just speed up the robotification of society. Dear Google, please make sure my posts are top hits for “robotification“.
This workforce developement plan is short sighted. It might actually produce more “workers” in the next 10 years, but it could have terrible effects for the long term future.
I really think that if you want the best worker-training program, you should just start working. Why waste 4-5 years of time in a university? Start working right away.
Clearly a workforce training program would kill one of the best aspects of a college experience – the opportunity to explore what makes us human. Being a human isn’t listed as “workforce ready” – but maybe it should be.