Quantcast
Channel: Co.Labs
Viewing all 36575 articles
Browse latest View live

Why Google Is Investing In Deep Learning

0
0

Google's acquisition of DeepMind Technologies last month was a huge deal. By snatching up the artificial intelligence company, Google signified a growing interest in deep learning. But what does this buzzword actually mean?

DeepMind was founded in 2012 by neuroscientist and former teenage chess prodigy Demis Hassabis and two colleagues. As its website describes, "We combine the best techniques from machine learning and systems neuroscience to build powerful general-purpose learning algorithms" with applications in a broad range of industries.

What Is Deep Learning?

Deep learning is an emerging topic in artificial intelligence. A subcategory of machine learning, deep learning deals with the use of neural networks to improve things like speech recognition, computer vision, and natural language processing. It's quickly becoming one of the most sought-after fields in computer science. But how did it turn from an obscure academic topic into one of tech's most exciting fields--in under a decade?

"Deep learning is being very highly prized at the moment," says Yoshua Bengio, full professor at the Department of Computer Science and Operations Research at the University of Montreal--home to one of the world's biggest concentrations of deep learning researchers. "The reason for that is that there is currently a lack of experts. It takes around five years to train a PhD student, and five years ago there weren't that many PhD students starting a career in deep learning. What this means now is that those few that there are are being prized very highly."

In the last few years, deep learning has helped forge advances in areas as diverse as object perception, machine translation, and voice recognition--all research topics that have long been difficult for AI researchers to crack.

Not To Be Confused With Machine Learning

To understand what deep learning is, it's first important to distinguish it from other disciplines within the field of AI. Early work in artificial intelligence dealt with explicit forms of knowledge--essentially, telling computers how to interact with their surroundings based on programmed facts and rules.

One outgrowth of AI was machine learning, in which the computer extracts knowledge through supervised experience. This typically involved a human operator helping the machine learn by giving it hundreds or thousands of training examples, and manually correcting its mistakes.

While machine learning has become dominant within the field of AI, it does have its problems. For one thing, it's massively time consuming. For another, it's still not a true measure of machine intelligence since it relies on human ingenuity to come up with the abstractions that allow computer to learn.

"A lot of successful machine learning applications depend on hand-engineering features where the researcher manually encodes relevant information about the task at hand and then there is learning on top of that," says George E. Dahl, a PhD candidate working in the Machine Learning Group at the University of Toronto. "The difference between that and deep learning is that the deep learning researcher will try and get the system to engineer its own features as much as is feasible."

Unlike machine learning, deep learning is mostly unsupervised. It involves, for example, creating large-scale neural nets that allow the computer to learn and "think" by itself without the need for direct human intervention.

"What the computer is learning using deep learning algorithms are more abstract representations of concepts," says Bengio. "Deep learning comes from the notion that as humans we have multiple types of representation--with simpler features at the lower levels and high-level abstractions built on top of that. By representing information in this more abstract way, the machine can generalize more easily."

Everyone Wants In On The Deep Learning Game

In 2011, Stanford computer science professor Andrew Ng founded Google's Google Brain project, which created a neural network trained with deep learning algorithms, which famously proved capable of recognizing high level concepts, such as cats, after watching just YouTube videos--and without ever having been told what a "cat" is.

Last year, Facebook named computer scientist Yann LeCun as its new director of AI Research, using deep learning expertise to help create solutions that will better identify faces and objects in the 350 million photos and videos uploaded to Facebook each day.

Another example of deep learning in action is voice recognition like Google Now and Apple's Siri. Much of this work owes a debt to Dahl, whose 2012 paper "Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition" represented a breakthrough in deep learning speech recognition.

"All recent speech recognition products by major companies either use deep neural networks of the sort that I work on--or else will soon," Dahl notes.

What is impressive is how dramatically deep learning can enhance these areas compared to the shallow networks and Gaussian mixture models (GMMs) used before. According to Google researchers, the voice error rate in the new version of Android--after adding insights from deep learning--stands at 25% lower than previous versions of the software. "In terms of speech recognition, we're going to see both wider adoption, and increasing gains in accuracy. That's where I think acoustic modeling is going," Dahl continues.

Thanks to deep learning, Yoshua Bengio says another area we're likely to see change in the next couple of years is the field of natural language processing. "This is something that companies like Facebook and Google are very interested in because the possibility of understanding the meaning of the text that people type or say is very important for providing better user interfaces, advertisements, and posts for your [news feed]," he says. "If deep learning can make the kind of impact in this area that it has in speech and object recognition, that could be a very, very important development in terms of value."

The Ethics Of Deep Learning

A unique development in Google's DeepMind acquisition was the mandatory establishment of an ethics board. According to people close to the situation, Google's willingness to establish an ethics board was a deciding factor in it purchasing DeepMind instead of Facebook. While almost any sci-fi movie of the past 50 years has dealt with ethical questions in some form or other, in the real world there are still relatively few concrete laws dealing with this part of AI--aside from the usual rules concerning things like privacy and product liability.

Bengio says this is with good reason: Currently the kind of models that can be built using even the most sophisticated deep learning tools are comparable only with the brain of an insect of terms of overall number of neurons. "Unsupervised learning is something that still presents big challenges--both computationally and mathematically," he says, explaining why fears concerning AI run amok may be a little premature.

George Dahl agrees. "We still have a very limited understanding of how the human brain works and some of that understanding might be 'platform specific' and not be relevant for artificial learning," he says. "Computers are a lot more powerful than they were even 10 years ago, but there's so much more scientific progress that needs to be made before we can realize the ambitions of researchers working in this field."

It's Still A Young Field

The call for an AI ethics board--and the resulting conversation--speaks less to where artificial intelligence currently is, and more to the level of public awareness surrounding it.

"We're nowhere near the kind of AI you see in science fiction, but that's not to say that deep learning isn't working in a whole lot of areas that are commercially viable--and which could be very useful to people," Dahl says.

A big part of what makes deep learning fascinating, says Dahl, is how fresh the field is.

"Computer science is a young discipline, and deep learning is a very young discipline within that," he says. "This isn't a subject like mathematics, where to make progress you have to be so specialized that only a few people can understand what you're doing. It's a young field--there's still a lot of low-hanging fruit, or else problems that may not end up being too hard, but which no one has yet had time to attack."

"It's very exciting for me to be working in a subject where there's so much potential to have an impact."


9 Reasons Being A Startup Founder Doesn't Suck

0
0

Over the last few years I've read a number of articles about why being a company founder sucks. While I can empathize, I think we need more public dialogue about why being a founder is worth it, not just long-term but in the moment too.

In our entrepreneur-obsessed times, this state of affairs is ironic. We know plenty about the inspiring skill and vision of our star founders: Zuckerberg, Jobs, Musk, Dorsey, Branson, Bezos. We have intimate knowledge of their financial windfalls. But we don't seem to know much about the day-to-day of a founder's life beyond a now well-touted list of challenges. I hope to whittle away a bit at this silence by illuminating some of the more rewarding things about company making.

In 2010 Brian Schechter and I founded HowAboutWe.com. HowAboutWe began as an innovative online dating site and has since evolved into a company with a suite of products designed to help people fall--and stay--in love. Our Brooklyn-based company of about 75 people has raised $22 million and is growing more quickly than ever.

Brian and I--best friends since childhood--started HowAboutWe knowing essentially nothing about company building. We learned everything about product development, marketing, and sales on the job. I think this steep learning curve has only amplified and intensified my founding experience--particularly the many wonderful things about this rather peculiar job. Here's my short list.

Constant Learning And Improvement

The most successful companies are the ones that are most effective at learning. Likewise for founders. Strong founders tend to be great learners. The other day a friend told me an inspiring story about watching Mark Zuckerberg grow as a leader while Facebook exploded; he described a young man who was constantly learning, constantly refining and iterating on his leadership style and studying the effects of the moves he made at the helm. This is in some ways the essence of company making: a crucible of perpetual, high-paced learning and improvement. For the curious-minded--those humans who thrive off attaining and applying knowledge--entrepreneurship is awesome.

Building a Culture Of Growth

A company is one of the most remarkable institutions for human growth. This is, interestingly, one of the least discussed aspects of companies and capitalism. I used to work in public service (education) and there, probably because there was very little money in the work, people talked endlessly about the personal benefits of their jobs. Much less so in business. But working at a company can be--and should be--an incredible growth experience.

In fact, I think one of the most important jobs of the founder is to create a culture in which each of your employees is having a life-enriching, perhaps even life-transforming, experience. This boosts morale, supports creativity, ensures retention, and increases productivity. It's also better for the world.

You'll Never Be Bored

Being a founder is not boring. And most founders hate boredom. There are 100 things happening at once. You are constantly shifting from context to context: from design review to strategy planning to recruitment to being on TV to fundraising to managing leaders. You need to be skilled at the big and the small, the human and the technical, the sale and the rigor, the drive and the execution. You won't sleep much. Your to-do list is absolutely insurmountable. If you find this fun, well, then it's fun.

Friendship

The bonds forged with your cofounder(s) and team members in the fires of company making are strong and long-lasting. There's a togetherness, a pride, and a no-bullshit mutual understanding that makes founder friendships glow. I've experienced this most clearly with my cofounder. When things are hard--and the path to success is inevitably hard--mediating challenges with another person can be exponentially more challenging than doing so alone. But if you can communicate well and stay open, if you can be honest and thoughtful, if you can avoid blame--then your relationship will flourish in the founding environment.

A Window

One of my favorite parts of running a company is the view it gives you into what's happening on Earth. I think this is particularly true for Internet company founders right now, since the Internet is among the most impactful human inventions since, I don't know, math. This knowledge of what's happening is a gift. It's a perspective that carries with it, I believe, a responsibility to stay humble and still figure out how to use your time and resources to genuinely move the lever for humanity.

Overcome Fear

Being a founder is an amazing opportunity to overcome fear--particularly the fear of failure. If you are a founder and are afraid of failing, it will be debilitating. You simply must find a way to overcome this fear. There are many stories that can lie behind the fear of failure: Perhaps you harbor some inner notion that perfection is required to attain worthiness, perhaps familial pressures or concerns about social status drive your fear, perhaps you are attached to material well-being, etc., etc., etc. As a founder, you must come to know what you fear and why. And, ideally, you will come to realize that the fear of failure is both deeply forgivable and always fundamentally delusional. All fear is.

Getting To Know Yourself

Building on my last point, founding a company is an incredible way of getting to know oneself. In my late teens and 20s I was a very serious meditator. Meditation is a profound tool for self-investigation. Starting a company is about as good.

Both activities--meditating and company making--provide you with a radical mirror. In both cases you don't have to look at yourself. You can float through. But maintaining ignorance almost takes more work than seeing things clearly. Your strengths, your weaknesses--they are staring you in the face all day, every day when you are creating a company. And, if you are willing to look at yourself honestly--with both self-acceptance and also an indefatigable drive to grow--then this is a true boon.

Creating Value And Knowing It

The purpose of a company is to create value. And in modernity that value is determined very simply by whether other companies or humans will buy what you're making. This simplicity is so honest. The result is that when you are creating something of value you feel that it's deeply real. There are so few places where value and meaning take on such an indisputable air. I like that. I think founders tend to like that. It's the feeling: I made this from nothing and I know that other people think it's worth something.

Founder As Career Path

Today, possibly more than at any time in history, large numbers of high-competency people are genuinely considering founding companies. This is likely a good thing as it fosters innovation, builds social value, and creates new jobs. It's also a fairly straightforward career choice. Some people decide to become engineers. Some teachers. Some founders. And yet, we know much more about what it's like to be a teacher than we know about what it's like to be a founder.

On the one hand, being a founder is a very public experience; we know a lot about founders in the same way we know a lot about celebrities. And yet, on the other hand--in a strange dichotomy--we know very little about founders, these private men and women who create and run our most influential institutions. We haven't yet quite learned to see the founder as a human.

And we haven't quite started thinking about the role of founder as a job, like other jobs, with its challenges and its boons. We should, though. Doing so would help existing founders and founders-to-be. And this, in turn, will help our companies, which I--along with many other founders--hope will help people everywhere.

The Fall Of Perl, The Web's Most Promising Language

0
0

I first heard of Perl when I was in middle school in the early 2000s. It was one of the world's most versatile programming languages, dubbed the Swiss army knife of the Internet. But compared to its rival Python, Perl has faded from popularity. What happened to the web's most promising language?

Perl's low entry barrier compared to compiled, lower level language alternatives (namely, C) meant that Perl attracted users without a formal CS background (read: script kiddies and beginners who wrote poor code). It also boasted a small group of power users ("hardcore hackers") who could quickly and flexibly write powerful, dense programs that fueled Perl's popularity to a new generation of programmers.

A central repository (the Comprehensive Perl Archive Network, or CPAN) meant that for every person who wrote code, many more in the Perl community (the Programming Republic of Perl) could employ it. This, along with the witty evangelism by eclectic creator Larry Wall, whose interest in language ensured that Perl led in text parsing, was a formula for success during a time in which lots of text information was spreading over the Internet.

As the 21st century approached, many pearls of wisdom were wrought to move and analyze information on the web. Perl did have a learning curve--often meaning that it was the third or fourth language learned by adopters--but it sat at the top of the stack.

"In the race to the millennium, it looks like C++ will win, Java will place, and Perl will show," Wall said in the third State of Perl address in 1999. "Some of you no doubt will wish we could erase those top two lines, but I don't think you should be unduly concerned. Note that both C++ and Java are systems programming languages. They're the two sports cars out in front of the race. Meanwhile, Perl is the fastest SUV, coming up in front of all the other SUVs. It's the best in its class. Of course, we all know Perl is in a class of its own."

Then came the upset.

The Perl vs. Python Grudge Match

Then Python came along. Compared to Perl's straight-jacketed scripting, Python was a lopsided affair. It even took after its namesake, Monty Python's Flying Circus. Fittingly, most of Wall's early references to Python were lighthearted jokes at its expense.

Well, the millennium passed, computers survived Y2K, and my teenage years came and went. I studied math, science, and humanities but kept myself an arm's distance away from typing computer code. My knowledge of Perl remained like the start of a new text file: cursory, followed by a lot of blank space to fill up.

In college, CS friends at Princeton raved about Python as their favorite language (in spite of popular professor Brian Kernighan on campus, who helped popularize C). I thought Python was new, but I later learned it was around when I grew up as well, just not visible on the charts.

By the late 2000s Python was not only the dominant alternative to Perl for many text parsing tasks typically associated with Perl (i.e. regular expressions in the field of bioinformatics) but it was also the most proclaimed popular language, talked about with elegance and eloquence among my circle of campus friends, who liked being part of an up-and-coming movement.

Side By Side Comparison: Binary Search

Despite Python and Perl's well documented rivalry and design decision differences--which persist to this day--they occupy a similar niche in the programming ecosystem. Both are frequently referred to as "scripting languages," even though later versions are retro-fitted with object oriented programming (OOP) capabilities.

Stylistically, Perl and Python have different philosophies. Perl's best known mottos is " There's More Than One Way to Do It". Python is designed to have one obvious way to do it. Python's construction gave an advantage to beginners: A syntax with more rules and stylistic conventions (for example, requiring whitespace indentations for functions) ensured newcomers would see a more consistent set of programming practices; code that accomplished the same task would look more or less the same. Perl's construction favors experienced programmers: a more compact, less verbose language with built-in shortcuts which made programming for the expert a breeze.

During the dotcom era and the tech recovery of the mid to late 2000s, high-profile websites and companies such as Dropbox (Python) and Amazon and Craigslist (Perl), in addition to some of the world's largest news organizations (BBC, Perl) used the languages to accomplish tasks integral to the functioning of doing business on the Internet.

But over the course of the last 15 years, not only how companies do business has changed and grown, but so have the tools they use to have grown as well, unequally to the detriment of Perl. (A growing trend that was identified in the last comparison of the languages, "A Perl Hacker in the Land of Python," as well as from the Python side a Pythonista's evangelism aggregator, also done in the year 2000.)

Perl's Slow Decline

Today, Perl's growth has stagnated. At the Orlando Perl Workshop in 2013, one of the talks was titled "Perl is not Dead, It is a Dead End," and claimed that Perl now existed on an island. Once Perl programmers checked out, they always left for good, never to return. Others point out that Perl is left out of the languages to learn first--in an era where Python and Java had grown enormously, and a new entrant from the mid-2000s, Ruby, continues to gain ground by attracting new users in the web application arena (via Rails), followed by the Django framework in Python (PHP has remained stable as the simplest option as well).

In bioinformatics, where Perl's position as the most popular scripting language powered many 1990s breakthroughs like genetic sequencing, Perl has been supplanted by Python and the statistical language R (a variant of S-plus and descendent of S, also developed in the 1980s).

In scientific computing, my present field, Python, not Perl, is the open source overlord, even expanding at Matlab's expense (also a child of the 1980s, and similarly retrofitted with OOP abilities). And upstart PHP grew in size to the point where it is now arguably the most common language for web development (although its position is dynamic, as Ruby and Python have quelled PHP's dominance and are now entrenched as legitimate alternatives.)

While Perl is not in danger of disappearing altogether, it is in danger of losing cultural relevance, an ironic fate given Wall's love of language. How has Perl become the underdog, and can this trend be reversed? (And, perhaps more importantly, will Perl 6 be released!?)

How I Grew To Love Python

Why Python, and not Perl? Perhaps an illustrative example of what happened to Perl is my own experience with the language.

In college, I still stuck to the contained environments of Matlab and Mathematica, but my programming perspective changed dramatically in 2012. I realized lacking knowledge of structured computer code outside the "walled garden" of a desktop application prevented me from fully simulating hypotheses about the natural world, let alone analyzing data sets using the web, which was also becoming an increasingly intellectual and financially lucrative skill set.

One year after college, I resolved to learn a "real" programming language in a serious manner: An all-in immersion taking me over the hump of knowledge so that, even if I took a break, I would still retain enough to pick up where I left off. An older alum from my college who shared similar interests--and an experienced programmer since the late 1990s--convinced me of his favorite language to sift and sort through text in just a few lines of code, and "get things done": Perl. Python, he dismissed, was what "what academics used to think." I was about to be acquainted formally.

Before making a definitive decision on which language to learn, I took stock of online resources, lurked on PerlMonks, and acquired several used O'Reilly books, the Camel Book and the Llama Book, in addition to other beginner books. Yet once again, Python reared its head, and even Perl forums and sites dedicated to the language were lamenting the digital siege their language was succumbing to. What happened to Perl? I wondered. Ultimately undeterred, I found enough to get started (quality over quantity, I figured!), and began studying the syntax and working through examples.

But it was not to be. In trying to overcome the engineered flexibility of Perl's syntax choices, I hit a wall. I had adopted Perl for text analysis, but upon accepting an engineering graduate program offer, switched to Python to prepare.

By this point, CPAN's enormous advantage had been whittled away by ad hoc, hodgepodge efforts from uncoordinated but overwhelming groups of Pythonistas that now assemble in Meetups, at startups, and on college and corporate campuses to evangelize the Zen of Python. This has created a lot of issues with importing (pointed out by Wall), and package download synchronizations to get scientific computing libraries (as I found), but has also resulted in distributions of Python such as Anaconda that incorporate the most important libraries besides the standard library to ease the time tariff on imports.

As if to capitalize on the zeitgiest, technical book publisher O'Reilly ran this ad, inflaming Perl devotees.

By 2013, Python was the language of choice in academia, where I was to return for a year, and whatever it lacked in OOP classes, it made up for in college classes. Python was like Google, who helped spread Python and employed van Rossum for many years. Meanwhile, its adversary Yahoo (largely developed in Perl) did well, but comparatively fell further behind in defining the future of programming. Python was the favorite and the incumbent; roles had been reversed.

So after six months of Perl-making effort, this straw of reality broke the Perl camel's back and caused a coup that overthrew the programming Republic which had established itself on my laptop. I sheepishly abandoned the llama. Several weeks later, the tantalizing promise of a new MIT edX course teaching general CS principles in Python, in addition to numerous n00b examples, made Perl's syntax all too easy to forget instead of regret.

Measurements of the popularity of programming languages, in addition to friends and fellow programming enthusiasts I have met in the development community in the past year and a half, have confirmed this trend, along with the rise of Ruby in the mid-2000s, which has also eaten away at Perl's ubiquity in stitching together programs written in different languages.

While historically many arguments could explain away any one of these studies--perhaps Perl programmers do not cheerlead their language as much, since they are too busy productively programming. Job listings or search engine hits could mean that a programming language has many errors and issues with it, or that there is simply a large temporary gap between supply and demand.

The concomitant picture, and one that many in the Perl community now acknowledge, is that Perl is now essentially a second-tier language, one that has its place but will not be the first several languages known outside of the Computer Science domain such as Java, C, or now Python.

The Future Of Perl (Yes, It Has One)

I believe Perl has a future, but it could be one for a limited audience. Present-day Perl is more suitable to users who have worked with the language from its early days, already dressed to impress. Perl's quirky stylistic conventions, such as using $ in front to declare variables, are in contrast for the other declarative symbol $ for practical programmers today--the money that goes into the continued development and feature set of Perl's frenemies such as Python and Ruby. And the high activation cost of learning Perl, instead of implementing a Python solution.

Ironically, much in the same way that Perl jested at other languages, Perl now finds itself at the receiving end. What's wrong with Perl, from my experience? Perl's eventual problem is that if the Perl community cannot attract beginner users like Python successfully has, it runs the risk of become like Children of Men, dwindling away to a standstill; vast repositories of hieroglyphic code looming in sections of the Internet and in data center partitions like the halls of the Mines of Moria. (Awe-inspiring and historical? Yes. Lively? No.)

Perl 6 has been an ongoing development since 2000. Yet after 14 years it is not officially done, making it the equivalent of Chinese Democracy for Guns N' Roses. In Larry Wall's words: "We're not trying to make Perl a better language than C++, or Python, or Java, or JavaScript. We're trying to make Perl a better language than Perl. That's all." Perl may be on the same self-inflicted path to perfection as Axl Rose, underestimating not others but itself. "All" might still be too much.

Absent a game-changing Perl release (which still could be "too little, too late") people who learn to program in Python have no need to switch if Python can fulfill their needs, even if it is widely regarded as second or third best in some areas. The fact that you have to import a library, or put up with some extra syntax, is significantly easier than the transactional cost of learning a new language and switching to it. So over time, Python's audience stays young through its gateway strategy that van Rossum himself pioneered, Computer Programming for Everybody. (This effort has been a complete success. For example, at MIT Python replaced Scheme as the first language of instruction for all incoming freshman, in the mid-2000s.)

Python Plows Forward

Python continues to gain footholds one by one in areas of interest, such as visualization (where Python still lags behind other language graphics, like Matlab, Mathematica, or the recent d3.js), website creation (the Django framework is now a mainstream choice), scientific computing (including NumPy/SciPy), parallel programming (mpi4py with CUDA), machine learning, and natural language processing (scikit-learn and NLTK)… and the list continues.

While none of these efforts are centrally coordinated by van Rossum himself, a continually expanding user base, and getting to CS students first before other languages (such as even Java or C), increases the odds that collaborations in disciplines will emerge to build a Python library for themselves, in the same open source spirit that made Perl a success in the 1990s.

As for me? I'm open to returning to Perl if it can offer me a significantly different experience from Python (but "being frustrating" doesn't count!). Perhaps Perl 6 will be that release. However, in the interim, I have heeded the advice of many others with a similar dilemma on the web. I'll just wait and C.

This Video Game Knows When You're Scared--And Gets Scarier

0
0

In the future, horror games will know when you're scared. And then they'll get scarier. Proof: the currently-in-development horror-adventure game Nevermind, which just launched a Kickstarter campaign last week. The game pairs classic first-person exploration with biofeedback data from a heart rate monitor in order to tell when you're scared and turn up the horror.

"In Nevermind, you get scared, you get stressed, and the world will punish you for giving in to those feelings," says creative director Erin Reynolds, "But it rewards you for calming down by becoming easier."

While biofeedback seems like a perfect fit for the horror genre, Reynolds believes that the technology is key to moving the video game medium forward as a whole, allowing for an entirely new level of immersion.

"I think it really speaks to the potential of games being able to know more about you than you know about yourself, and having this intimate response to your internal reactions," Reynolds says.

That internal response surprised her during playtesting, as it illuminated "just how personal one's sense of horror is. It made for some design challenges, because it means you need to have something for everything so that everyone's buttons get pushed."

But those challenges also served as the ultimate affirmation for Reynolds: She was scaring people.

That's a good indication that Nevermind may be a successful game and not just a neat tech demo. Reynolds has ambitious goals for the game and hopes that it will move the medium forward as a proof of concept in both biofeedback integration and as an example of a positive game that reinforces stress management skills that have real-world applications.

Because achieving those goals with a video game is all for naught if the game is not fun, states game developer Lat Ware in a feature on Gamasutra:

"The best practice in making biofeedback games is also the best practice for game development in general: Make it fun," he adds. "Fun is the only thing that matters in a game. Fun is what makes people love your game. Fun is what makes people come back to play again. Fun is what makes people buy your next game without asking questions."

"That's why I'm really excited about Nevermind," says Reynolds. "It creates this experience that is fun but can also empower the player."

How The Founder Of WeTransfer Went From Blog To Startup

0
0

If one were to compare the Dutch tech entrepreneur Nalden (yes--he goes by just one name) to anyone in the States, it would probably be a hybrid of the serial entrepreneur Sean Parker and Jason Kottke, the web designer and blogger. Like Kottke, Nalden gained a large following and livelihood by blogging about stuff that he liked--mostly music and DJ culture--and similar to Parker, Nalden is an idea man who is uncovering the hidden possibilities of the digital world faster than most. But he has also become a brand of his own, able to sell big-label products just on his credibility as a style maven.

Nalden

With blogging now behind him, Nalden is breathing life into his two businesses, one of which you've almost certainly heard of: WeTransfer, a site that makes sharing large files over the Internet easy, and lets users pick high-quality background images for the sharing page. His other company, Present Plus, brings art and technology together in the public domain.

He started as a teen blogger who blogged to let his friends know "what he thought was cool" but ended up becoming a country's cultural ambassador. When actors like Clive Davis and Sacha Baron Cohen would come to Amsterdam, Nalden would show them around, thanks to his PR connections. His eponymous website became prime real estate for big-name advertisers like Apple and Nike. He was the eyes and ears of a generation of digital natives, with his influence spilling over Dutch borders.

Nalden, whose real name is Ronald Hans, used his childhood nickname to baptize his blog, Nalden.net. The professional pseudonym stuck.

In a 2009 interview the manager of the singer Anouk told the magazine Vrij Nederland, "Nalden knew how to give the zeitgeist a good spin and played a pivotal role in the [hip-hop] scene. With a fresh, fun voice."

His voice attracted promoters who wanted in on the Millennial demographic. "My site became profitable within six months [of starting it]," he said. Nalden was just 16, and the year was 2002, "before blogging was a thing."

When Nalden shut his site down in 2012, he received a ton of mail. "One person even wrote me to tell me that he chose his job because of my site. It was amazing. I mean all you see is numbers, but you never meet the people," says Nalden.

Perhaps the most notable part of Nalden's advertising scheme was his Wallpaper Model. Sick of seeing every website use unattractive banners to push ads to online users, he was bent on bringing the beauty of glossy magazines to the Internet.

The 29-year-old's signature Wallpaper Model has made its mark in Nalden's more recent projects. WeTransfer uses "the same DNA" as his blog, with artistic ads and outright artwork gracing its transfer page. "There's no clutter. We get a click-through-rate, on average, of 2%. Most banners get on average 0.014%," he says.

And one of Present Plus's current projects, Kuvva, makes tons of curated wallpapers from handpicked designers and illustrators available to mobile and desktop users.

When I met Nalden at Present Plus's Amsterdam office, we took a moment to admire the art pieces that his team curated in their first-floor gallery. "We just finished a show. It was about dinosaurs," he says.

Somehow, the innocence of the dinosaurs made me realize why everyone wants to get to know Nalden. It's really simple things that we want to surround ourselves with. Coincidentally, it's also his design philosophy.

Becoming A (Technical) Blogger

Nalden grew up in the small town of Wilnis, the Netherlands, and started making websites for his dad when he was 13. "My dad is a conductor. So he didn't really need a website. But it was so cool to me that I could publish this thing and everyone in the world could see it."

He recalls his first computer that ran Windows 3.1. "I used Microsoft Frontpage and basically taught myself. I still have this website on a floppy disk at home. The first page was just text, pictures, and a treble clef."

Nalden's parents never really pushed him to become an entrepreneur, but he had the drive to learn how to improve his site and create new things. When he created his blog, Nalden says he "coded every day to tell a short story," using WordPress as a foundation.

In the very beginning, he frequently blogged about music. "I was always illegally sharing music with everyone on my site. This was the very early blogging era. It was Universal Music. They basically said, if you don't stop doing this, we will sue you. But I introduced a new business model for them." They promoted their music on Nalden's blog after that.

In the 10 years that followed, he took on major advertisers and relocated to Amsterdam. The design firm Momkai approached Nalden in 2007 to collaborate on his blog, and the Wallpaper Model was born.

"The idea was to do a different layout for a blog, based on a desktop model, taking Mac OS X as a reference and at the same time push beautiful wallpaper images while the content floats on top of it," Nalden says. He let Momkai handle the coding and maintenance of his site from then on.

Exposure to all of these brands whetted Nalden's appetite for entrepreneurship. With a couple of friends, he started WeTransfer, aiming to keep it user-friendly. "I told myself the baseline was my dad. He had to be able to use it," says Nalden. They now have around 20 million active users per month.

Originally, WeTransfer hosted all of the data on their own servers, but their user base was growing so rapidly, and is still growing at 1 million new users per month, that they eventually moved to the cloud and have Amazon Web Services host. "That way, I don't have to hire any engineers to build server parts. We want to focus on the user experience," Nalden says.

Half of WeTransfer's revenue goes into supporting the artistic community and now exists as a separate entity from Present Plus.

Merging Art and Technology

Three years ago, Nalden met Damian Bradfield, who was running an advertising agency on the other edge of Amsterdam.

"We talked about how we hated everything and wanted to change it," recalls Bradfield of when they first met. "I decided to go into business with him. Advertising was dying anyway," he says.

Present Plus employs 16 designers, developers, and creators with offices in London and Amsterdam. They create video, illustrations, photos, editorial content, and a hosting platform for their clients' work. "Most sites use photos to display their work. They look like photographers. Video really shows what they did. And bloggers will embed those videos. It works," says Nalden.

Present Plus prides itself on taking care of all of a clients' advertising needs. "Companies are really, like, 'Ha! I did my job!' And then we do the video, coding, everything," says Bradfield.

But Nalden insists on producing original ideas. "Present Plus was never made to do client work only. We still want to make our own products," says Nalden.

Instead, Bradfield and Nalden are pushing Present Plus in a new direction, bringing young artists' work to more audiences through technology. Their new program, called Artists & Algorists, is exploring the future of digital art. For it, the Present Plus team is actively cultivating a community of artists with experience in coding and digital interaction. Their creative lead, Sam van 't Oever, has created a handful of original pieces using WebGL and HTML5.

"I think my generation and all the entrepreneurs who are starting businesses grew up with the Internet, and so in terms of being creative, you sort of develop yourself as a different creative person. If you know how to code or how to design a website or at least know Photoshop and know how to transform that Photoshop design to a stylesheet that's, in my opinion, a new way of being a creative person," says Nalden.

Their flagship Artists & Algorists project is now an interactive installation in the lobby of the Art'otel Amsterdam. Designed by Present Plus's club of artists, onlookers have a physical back-and-forth with the projections while their movements are captured with Microsoft Kinect technology.

"When you dive into coding and dig deeper into the code, then you become aware of the possibilities that are out there, and that's led to a whole new generation, the next level of creative people," Nalden says.

Why "This Is My Jam" Stopped Overthinking Music Discovery

0
0

The question will likely be timeless: How do you discover new music? Some companies have bet on human editors, while others put their weight behind clever algorithms. One company thinks the solution is actually more like a pendulum that swings back and forth between the two.

"The best music discovery happens in a context," says Matthew Ogle, founder of This Is My Jam. "It comes from a friend, from someone you admire and trust, or at a specific time or place." The discovery question is one that Ogle has been trying to solve by asking a slightly unusual question: Which song is most likely to hook you on a new sound and keep you listening?

That One Song

That One Song is a new feature from This Is My Jam that tries to determine that must-hear song by a given artist. It takes all the aggregate data about their songs and narrows the choice down to one track. To date, there have been around 1.5 million songs posted to the company's site, of which about 500,000 are unique--not duplicates posted by different people.

Visiting the one song page, you literally fill in the blank. "What should I listen to by ____?" The feature was born quickly in about 48 hours and was spurred on by the Midem Hack Day. Searching through the concert listings in Cannes--where Midem was being held--and not recognizing many of the artist names led to the experimental feature.

Behind the scenes the feature works by grouping all the data and then regrouping by artist. Once that's sorted it's then weighted not only by raw number of times a song has been mentioned as someone's jam, but also adds play count and number of likes into the algorithmic decision as well.

"This method worked surprisingly well out of the gate," says Ogle. "Often highlighting either the obvious hit like "Royals" for Lorde, but also frequently throwing a song into the #1 slot that is well known but not usually on top, but still has that certain "jam-factor" like The Beach Boys' 'God Only Knows.'"

What about the obscure songs that have a lot less data attached to them? When someone picks those, it gets assumed that you're a bit more of a music connoisseur and your choice gets weighted heavier. It's an aspect that has more of the human element behind it. The one-song feature is a first step for the company that's finally able at a user level to see patterns in the data.

"The data is now in good enough shape that we can build public-facing experiments on top of it, like That One Song," Ogle says. "We're super excited to light up those, currently invisible, connections between songs and people, and are having a lot of fun figuring out how best to bring those to life within the context of the site."

For those that follow the music tech scene, it may seem a bit odd for This Is My Jam to being doing its own data mining. The Echo Nest is a company known for making connections between various types of musical data that everyone from Twitter to the BBC use, but interestingly, This Is My Jam chose to spin off from The Echo Nest and is doing all its own data work.

"I was working on a bunch of different projects for the The Echo Nest at the time and This Is My Jam was initially a 'side project' idea that they, to my pleasant surprise, saw the possibilities in and helped support," Ogle says. "By the time the site started taking off, the The Echo Nest's own B2B (business-to-business) business had started to catch fire as well, and it didn't make a ton of sense for them to be incubating us forever so we decided to spin it out. An unorthodox route to an independent startup but it worked for us."

This Is My Jam

This Is My Jam still uses The Echo Nest to ensure that artist names, track titles, and other metadata are correct.

Just because This Is My Jam started as a simple one-trick pony doesn't mean it'll always stay that way. It also doesn't mean that Ogle thinks building a simple service is the solution.

"It's not simplicity for its own sake--though we do think simple experiences are delightful," Ogle says. "But it turns out simple experiences are great building blocks over time for more complex products/services."

Can This Is My Jam answer the tired question of how music discovery should be done? It doesn't hurt that Ogle has a solid grip of the actual problem outside of buzzword territory. "Making 'discovery' the end goal of a service can be tricky--we struggled with this at Last.fm for a while," says Ogle. "It's actually quite rare that someone sits down and says 'I would like to discover music for the next 35 minutes.' So it becomes more about asking 'What existing behaviors around music can we enhance?' or 'How do we motivate people with great taste and reputation to share the love." Do those sort of things well and discovery will be what happens incidentally in the cracks."

The NBA's Development League Straps A Sensor Disc To Every Player

0
0

Your favorite basketball player is about to get one step closer to being a cyborg.

The NBA's Development League (D-League) will soon begin experimenting with wearable technology on the court, the league announced today. A small disc weighing in at a whopping one ounce--attached either to players' chests or between their shoulder blades and worn underneath their uniforms--measures vital biological statistics.

Developed in conjunction with STAT Sports, Catapult, and Zephyr, this groundbreaking wearable tech makes available--in real time--individual players' current state and statistics. The information is relayed to coaching and medical staffs alike in an effort to improve players' efficiency and effectiveness on the court.

"Our goal is to help improve our players' performance and maximize their productivity on the court and help them live their dream of someday playing in the NBA," says D-League president Dan Reed.

By better understanding a player's cardiovascular and musculoskeletal status during games, the technology promises to quantify exertion in a number of categories including: acceleration and deceleration, jumping, distance, and direction run.

As of now, just two teams are known to be using these little buggers: the Bakersfield Jam and the Ft. Wayne Mad Ants, but the league hopes to expand well beyond that in the future, hopefully by the end of the season, a goal Reed likens to the NBA's recent partnership with STATS VU and SAP, and incoming commissioner Adam Silver's "commitment to the technological advancement of the league."

This Firefighter Built His Own Google Glass App And It's Saving Lives

0
0

Some of the veterans in the Rocky Mount, North Carolina fire department think he's crazy, but that's not stopping Patrick Jackson. The self-taught computer programmer and full-time firefighter isn't fazed by sideways looks he gets walking around the station wearing Google Glass. That's because he knows it could save lives.

Jackson, a 34-year-old member of the Google Glass Explorer program, has developed an app that displays incoming emergency dispatches, shows maps of where incidents are, nearest fire hydrants, and even building plans. By using data to shorten response times, he's hoping to turn his peers' disbelief into respect.

Patrick Jackson

The firefighter's savvy creation is part of a growing subculture of social good apps that could turn Glass into a tool to save lives. In Michigan, a young IT professional is using Glass as an alert system to help sleepy drivers stay awake and direct them to nearest rest areas via GPS navigation. Meanwhile, doctors at the University of Alabama, UC San Francisco, and UC Irvine are even using Glass to help perform surgery.

Glass's greatest potential lies not in fashion statements, or even in a new way to share social updates and capture media, but in helping professionals work more efficiently and safely. It can even improve the quality of life for the disabled and help children learn.

Glass For Emergency Services

Jackson, who learned to code at the early age of 7, studied engineering in college before transferring to environmental policy management. After a 10-year hiatus from computers, Jackson was inspired to develop Android apps, creating the app Firefighter Log for smartphones around 2010. Integrating the department's dispatch system, facilitating instant notifications, closest hydrants, and maps, the department has used the tool for the past two years, with firefighters downloading it across the country.

Last year Jackson heard about Glass, and raised enough funds via Indiegogo to get a pair of the $1,500 frames.

Jackson's Glass app allows for all incoming emergency dispatch calls to come directly to his device. He then uses Glass to lead his unit to the location and find the nearest fire hydrant while en route, saving precious time.

When someone calls 911 and asks questions from dispatch, that info drops into a database which the app picks up. Jackson then gets a notification, often while the dispatcher is still gathering info. He says this gives as much as 30 seconds to a minute of lead time to act, "which could be big in a fire or a heart attack."

So far the department is only testing this out--it hasn't been used at the scene of an emergency. But Jackson does use the app at the station and while traveling, so far using Glass on about 100 calls over the past several months.

"It gives me a real quick way to get notifications, as well as a map of where it is, along with notes that the dispatcher can type in," he explains, wearing it on the way to incidents. "I won't wear it on EMS calls because people don't know what it is, and its an invasion of privacy having a camera there."

Now Jackson is developing Glassware to allow him to look up location information, and structural floor plans by saying the address or just looking at the building. In a new spin on the hydraulic rescue tool, jaws of life, Jackson is integrating car diagrams in case of emergencies to cut through cars, allowing firefighters to easily find the best access points.

Jackson says he thinks Glass will soon have a widespread role in the field with EMS. One use could be recording video after leaving the station and recording upon arriving at a scene, useful for arson investigators and for training new recruits.

"I see Glass as being very helpful there in a hands-free way to get information." Having Glass hasn't saved a life, yet, but it has saved property, he says. "Given the right situation it could."

Introducing the tool to the other 150 firefighters at the station wasn't easy for the seven-year veteran and ladder truck driver. But he's starting to win them over.

"Most of the guys think it's a good idea, but I get picked on for how it looks," he admits. "First day I came back with them there was actually a guy who made a big pair of safety glasses, taped cardboard on it and wrote 'Google' on the side."

But since a YouTube video was released a few weeks ago, showing the app in action, things are changing. The video has amassed almost half a million views, causing "a lot of response from other firefighter and departments."

Jackson is also working on a CPR assist app for Glass, measuring the speed of compressions, and whether you need to speed up or slow down based on sensors that detect head movement. He's teaming with a Michigan startup called team(evermed) during his days off from the department, where he spends 10 days per month working grueling 24-hour shifts.

The fire engineer is also working on an electronic medical application to display patients' records on Glass.

"I'm looking at making this a company and making it available to other fire departments, and making it affordable and scalable," he says. Because of liability issues, "it may be a training aid or used in health care providing seeing or first responders."

With a few minor tweaks, Jackson thinks these tools can be extended to police and other emergency medical services.

"It wouldn't be that much of a jump," he claims, and he's exploring those options right now. The firefighter turned entrepreneur plans to either bootstrap his business and work at the fire department while getting customers, or seek investments and leave to pursue the project full time.

He envisions Glass apps like his as a future standard service tool for firefighters. In fact, he has already been making the rounds at industry conferences like SMART firefighting, a CES-like event for mobile devices and emerging fire service tech, and there's interest from other departments.

"I think its going to happen," he says. "It'll be a few years. The hurdle right now is the price."

Glass Behind The Wheel

Only two years removed from college, Jake Steinerman spends a lot of time driving up and down the state of Michigan, putting 24,000 miles on his 2013 sky blue Mazda in search of prime snowboarding runs. Trips have included some long nights driving, and more than once, an exhausted Steinerman has fallen asleep behind the wheel.

About eight months ago Steinerman says he passed out for five to 10 seconds while coming back from Detroit around 3 a.m. Luckily the New York native woke up and got off the road before tragedy struck. The experience shook and inspired him to create a tool to help overextended and exhausted drivers. It's called DriveSafe, an app that prevents you from falling asleep while behind the wheel. Using Glass' sensors, DriveSafe can detect when you doze off via built-in infrared sensors.

"It's like having an invisible driving assistant, something that's always there and always watching, like a friend sitting next to you," Steinerman explains. "The app looks at motion of your eye and the tilt sensor. When you reach a certain threshold, you hear audible feeds."

When that happens, DriveSafe triggers an alert to wake you up and a text notification display saying you're going to sleep. When you tap on Glass, the app then uses GPS navigation to direct you to the nearest rest area so you can continue driving safely later.

"It's not going to shock you," Steinerman clarifies. "It's going to ease you so you don't overcompensate and go into an accident for that reason."

The 23-year-old IT professional wants to turn this labor of love into a full-fledged business. "Right now our starting point is sleep detection and drowsiness alerts. We think we can tackle issues like distracted driving, using a compass to tell if you're focused on the road, halfway turning your head around, or if you're looking at, and replying to text messages and emails."

Steinerman thinks his hands-free device is far less distracting than smartphones or dashboard GPS. Real-world applications go far beyond late night drivers, including "industries like delivery, truck drivers, and machine workers."

The 2012 University of Michigan grad previously spent seven years volunteering with autistic children in Long Island, New York. That experience inspired his first startup, a now defunct speech help company called MoBlue Tech. Even though he wasn't able to scale that business, Steinerman considers that time instrumental in his development.

"It's great to see folks be part of the Glass community and be passionate, not just create an Angry Birds for Glass, but an app that really will change lives," Steinerman says. "It really allows people to do things they weren't able to do before. That's really where tech is needed. The first time I think in a long time we see tech that's really different and built for science."

Other Innovations To Watch For

Right now apps are being developed to help the deaf, the blind, and by surgeons in the U.S. and overseas--things like Navatar, a Glass app built by the University of Nevada at Reno for the blind that gives audible directions in indoor spaces.

Firefighters and doctors are already benefiting, but it's likely not long until other professionals, and even blue collar workers like truck drivers, plumbers, electricians, manufacturers, and factory workers, tap into its benefits.

None of this would be possible without Thad Starner, the Georgia Institute of Technology professor who coined the term "augmented reality" back in 1990, and helped developed Google develop Glass.

Starner, who is involved with a slew of Glass-centric innovations, has been sporting wearable tech for over 20 years. He says this wave of developers is just the beginning. And that's in large part thanks to those explorers playing with early forms of the device, and unlocking possibilities which may influence future Glass' functionality, "exactly what we're seeing from firefighters and doctors. We couldn't do that with a mass-market device," Starner says.

Right now students at Georgia Tech are working on an app similar to the technology that the Marvel comic book superhero Mr. Fantastic used to decipher unknown languages back in the 1960s and '70s. Except this is science fact--not fiction. And it's not a gun that zaps you. It's called Translator on Glass, and it could have tremendous value in places like hospitals and emergency situations.

"It's very hard at 2 a.m. to find yourself an interpreter," says Starner, who knows firsthand the value of that tool, raised by a mother who worked late nights in the emergency room of a New York hospital.

Here's how Translator on Glass works: A doctor or nurse would hand a patient a phone, they would select their native language, tap on the device when they wish to speak, and talk into the phone's microphone in their native language, which then "ships it off to Google translate, and then sends it to the doctor's eyepiece in English," Starner explains. "The doctor can tap his Glass and speak in English, and Spanish (for example), shows up on the cell phone. Now we have an emergency translator," which could be used beyond medical professionals, helping EMTs, firefighters, and police communicate.

Another project worth watching is Captioning on Glass, by Jim Foley, a Google scholar and computer science professor at Georgia Tech. Foley has been progressively losing his hearing, and he created a tool to specifically help other sufferers.

People with impaired hearing often have to focus on people's lips and body language to understand conversations, which can be very hard to pick up in certain environments. Foley's app allows users to tap Glass whenever they want to speak, which then transcribes their words into a phone and sends it as text on Glass.

"It allows you to actually pay attention to other people's mouth and facial gestures," says Starner.

What's Next For Google Glass?

With Google planning a mass consumer launch later this year, subsequent iterations will follow, further opening up the options for developers. Starner is quick to remind that we're still in the very early days of large scale wearable technology, and prices will drop as Google ships and competitors hit the market.

"This is like the first mobile phone," he says, which cost around $2,000 in 1982. Back then, much like Glass, it was used by doctors and lawyers. "Then as the technology developed, it got cheaper."

Starner compares Glass to the PalmPilot in the 1990s, predicting we will soon see a "fast adaptation of the device to all sorts of things."

"Explorers have seen and hacked the thing to see the future for it."


How This Startup Designed A Printable Bluetooth Keyboard

0
0

The next time you need a new keyboard, you may be able to use one printed on paper. Novalia, a U.K.-based startup, recently announced a super-thin, Bluetooth-enabled keyboard that can be printed on paper. For Novalia founder Dr. Kate Stone, the outcome wasn't a triumph, but a compromise. "Identifying compromises and [letting] those compromises affect the design of the product itself," is what guides her team, she says.

Novalia's Design Methodology

Stone hinges her entire design theory at Novalia on an old saying from her grandmother: "Cut your coat from the cloth at hand." In other words, use what already exists to reach the next step of transformation.

Because of this approach, the team finds themselves turning toward generic, unexciting objects as a jumping-off place from which they can discover innovative technology. For instance, when it came time to decide what type of ink to use in the control board, the decision wasn't a trivial one.

"We don't use the most conductive of inks," Stone says. "We use inks that work on a label press. This influences our choice of electronics and then the software we create."

How A Printable Keyboard Works

"For several years we have been exploring how to add interactivity and connectivity to printed items using traditional print," Stone says. Eventually, one team member was able to write the code for something that would be both Bluetooth and touch enabled, a fundamental aspect of the current keyboard's features.

The actual keyboard, currently just a prototype, will be printed on industrial printers using a Nordic Semiconductor chip, which houses the Bluetooth and capacitive sensor tech. The Cambridge-based Novalia created "two type[s] of printed touch, one is segmented, discrete touch points, the second is a touch matrix, where touch points are defined in software," Stone says. The only active part on the keyboard is the ink.

Through coordination, the capacitive sensors communicate via the chip and Novalia's own software that runs on Bluetooth. For easy integration with other products, all of the technology is contained in a sheet that is only a few millimeters in thickness.

"We really believe in connecting everyday objects to smartphones and tablets, and creating low-cost ways of extending ways to input to these devices," Stone says. For instance, the hardware could be embedded within the pages of a book and once someone touches a certain region the page could play music or an app could consequently open on a smartphone or tablet.

The Five Secrets To Running A Totally Distributed Company

0
0

Steven Schuurman is the CEO of Elasticsearch, which provides real-time search and analytics tools to developers based on the Elasticsearch, Logstash, and Kibana open source projects. It was a distributed company from day one, partly because of its open source roots. The projects' contributors and users were already scattered all over the world.

"We said we want the best people period," says Schuurman. "We just don't care about their location. The best distributed organizations completely and utterly embrace the fact that they are distributed. If you can get the distributed nature of an organization to work for you, you also get infinitely better interaction with your install base."

Elasticsearch has two main hubs--in Los Altos, California and Amsterdam in the Netherlands--and a slew of smaller offices in places as diverse as Phoenix, Paris, Prague, Austin, Boston, Barcelona, Berlin, and Romania. Two thirds of its employees are developers.

Elasticsearch is not Schuurman's first distributed company. He was also one of the seven founders of SpringSource, a company built on the open source Spring framework which was sold to VMware in 2009 for $420m. Several of SpringSource's founders had never met in person before they started the company together.

"Messing something up is much easier in a distributed org," says Schuurman. "In an ordinary company, you can make many more mistakes because you spot them earlier." Here's how to avoid some of the those mistakes.

Be A Baby Multinational

Setting up an international commercial organization is complicated and expensive. You have to deal with different legal systems, tax regimes, accounting rules, and language barriers. "Those are way harder to deal with than when effectively you just communicate about code and about product," says Schuurman.

You'll need an accountant who specializes in international accounting and a law firm which has a global network. One of the first people Schuurman approached to join Elasticsearch was its CFO, Nick White. White set up all the legal entities, and commercial and reporting infrastructure for the new company within a few months.

"Find people who can advise you how to get it done properly," says Schuurman. "It's a massive investment and a big cash drain but you do it anyway because if you don't get it right in the first year or two and you start scaling up the organization, you are in a world of hurt later on. You have to get it right pretty much from day one."

Find Demonstrative Developers

Executives at Elasticsearch must work at one of the main hubs, partly because they need need to interact regularly with the company's investors and board of directors. "Also, the simple truth is that there is the highest concentration of A+ quality IT execs in the Bay Area that can run high potential startups," adds Schuurman.

Engineers, on the other hand can be located anywhere, which is not to say that every developer is cut out for this kind of environment. Elasticsearch looks for people who are smart and skilled but also highly communicative.

"People that are fairly outgoing, that won't chew on a problem for days without reaching out," says Schuurman. This also means that problems are highlighted early. "Before an issue becomes a real problem, it's on the table. It's out in the open."

More importantly for Schuurman, his distributed employees need to make a broader contribution to the company, often taking advantage of their location to do so. "There are guys who are really, really smart who we just won't hire. We want people who are interested in more than just the code."

"That doesn't just apply to engineers. My sales guy that sits in London enables the London Elasticsearch user group. He looks beyond his direct responsibilities. I don't care whether it's my accountant in Germany or my sales engineer in Portland or my VP of sales in Los Altos, these guys all care about how do we work together to make our customers happy."

Then Set Them Free

Once you have hired the right type of engineer, let him manage himself. "If you are going to have someone who sits 1,000 miles away from his nearest colleague," says Schuurman, "you'd better get someone who is super comfortable with operating like that, who has the freedom to take his kids to school in the morning, does grocery shopping right after, and only starts work at 10:30 a.m. You have to give people the possibility to completely embrace the fact that they have a lot of freedom: how they go about their process, how they manage their time."

Finding the right developers regardless of where they are and giving them the freedom to decide how they work seems to be paying off for Elasticsearch. "I've never seen a team this productive," says Schuurman. "It's shocking how productive this team is. We have been able to take Elasticsearch from a project to a thriving company in just over a year at four times the rate at which we grew at SpringSource."

Be Crystal Clear

According to Schuurman, one of the biggest shifts in going from an open source project to a company-based is that "you have to make sure that whatever you do is more predictable and reliable." The first step to achieving that is making unambiguous decisions and communicating them clearly.

"At the end of the day you're not sitting across the room from your direct colleagues," says Schuurman. "Sending out mixed signals or setting objectives that are subject to misinterpretation can easily lead to folks working toward the wrong goal."

"The way I have always approached this is by repeating my goals and objectives over and over again plus I constantly double check whether they are well understood by discussing with folks how their deliverables contribute to the higher objectives. Whereas this is especially critical in a distributed org, it's really a best practice that all companies should follow."

Face-To-Face Time

"A big pitfall for a distributed organization in today's world of Skype and Google hangouts and what have you," says Schuurman, "is that you forget about the value of face-to-face. We heavily invest in face-to-face time."

The entire company gets together about twice a year but Elasticsearch also takes every opportunity to bring employees who are not in the same office together. "When we do a public training course, we never have just one teacher, we always have two. We ideally never send one person to a conference even if we don't have a booth or just have one talk. It costs a fortune. At least it looks like it costs a fortune, but it's worth it. You build much better rapport."

A lot of Shuurman's advice boils down to being extra-disciplined. "You have to be so critical. Your standards have to be super high in order to really make it work successfully," he concludes. "The knock-on effect of that is that you have an organization that becomes run so efficiently and becomes so productive that it shows in your numbers."

10 Steps To Pulling Off A Killer Hardware Hackathon

0
0

As anyone who has organized one will tell you, organizing hardware hackathons is notoriously hard to get right. In many ways hardware hackathons are more "advanced" from an organizational standpoint than their software cousins are because the nature of hacking hardware requires more tools and parts than the standard laptop, Ethernet cable, and Internet connection attendees at a software hackathon need.

But a good hardware hackathon is so much more than the availability of tools and parts. It's about the experience, the energy, the space and lighting, and even the beer. So how do you get one right? I talked to two experts to find out.

Remember, You're Just The Janitor

Before we jump into the step-by-steps of throwing the best hardware hackathon the world has ever seen, there are two very important things you as the organizer need to remember. The first: It's not about you, it's about your hackers.

"Our role as organizers is to be the janitors, the line cooks, and the gardeners," says Zak Homuth, cofounder and CEO of Upverter. "Our entire job and our only responsibility is to enable these hackers. Get them to show up. Empower them to hack. Give them a reason to stay. Help them to communicate. Feed them. Fill them with sugar and booze. Bribe them with other smart people. Promise them you won't do any of the shit they deal with at work. And stay out of their way."

So while the hackathon may be all your idea, the star of the show will always be the hackers. That also goes for people who are organizing hardware hackathons for big clients who want to see what hackers could do to manipulate, change, and improve on their existing hardware products. In that case the client's hardware product may be the star at your client's company, but at a hackathon--even one dedicated to hacking only that piece of hardware--the product takes a backseat to the hackers.

The second important thing you need to grasp before you even start planning your hackathon is that it is a hackathon you are throwing and not a workshop. The distinction between the two is nuanced, but critical, and not realizing how the two are different is what has led to plenty of bad "hackathons" over the years. So what's the difference?

"I think this question is easiest to answer from the perspective of an attendee," says Andrew McWilliams, who is the brains behind the Hardware Hack Lab at ThoughtWorks NYC. "If I attend a workshop, I expect a curated experience in which I am guided by subject domain experts. I expect to proceed in a structured way, using exercises that are designed to build on each other progressively to achieve a stated overarching goal or goals. If I attend a hackathon, I expect some kind of orientation and then to be let loose. I may seek guidance from an expert along the way, but I am more likely to come across an innovation or insight serendipitously. I am aware of the general aims of the hackathon, and I and my team are open to design and change our approach as we go."

If you start organizing your hardware hackathon with these two things in mind, than the rest is relatively easy as you'll be approaching the following steps in the right frame of mind.

1. Figure Out Your Aims, But Be Broad

This is the only part of the hackathon planning where you are more important than you hackers. Throwing a hackathon isn't like throwing a party. The point matters. And the only one who knows the point (at first) can be you.

Both Homuth and McWilliams agree that organizers must have a clear purpose of what the hackathon is for, but it is important not to limit your hackers' creativity by giving them specific instructions on what to do.

"Wherever possible, be non-prescriptive," McWilliams says. "For example, 'Explore possibilities between RGBD and Oculus' is better than 'Write RGBD shaders for Oculus.'"

Once you know your aims, plan a hackathon that is long enough to allow hackers to produce results. Is 24 hours enough? Or would a weekend or an entire week be better?

2. Build A Community

Once you know the broad point of your hackathon, start building a community around it. How will participants find each other and communicate before and after the hackathon? How will they communicate with you?

Hackathons work by allowing communities of like-minded creatives to bounce around ideas and work off each other's vibes. Allowing the attendees to easily communicate with you and one another before the event will pay off in spades later.

A good way to find motivated participants is by partnering with external organizations. "ThoughtWorks encountered a notable rush of new energy at the Hardware Hack Lab when we partnered with the Volumetric Society of New York," McWilliams says, and also notes that it's a good idea to think about the subject-matter experts who may be able to offer guidance to groups, and what questions might be asked by participants.

Another simple tip for building community?

"Hand out T-shirts," McWilliams says. "It works!"

3. Plan Hardware And Resource Requirements

Step three builds directly one steps one and two. The better you know your aims and the more you can build a community that can openly communicate with you and each other before the event, the better you'll be able to provide for and plan hardware resources and requirements.

"To the extent that it is possible--remember we are trying to be non-prescriptive--figure out what type of groups might form and what hardware and resources they each may need," McWilliams says. "Ask participants what their motivations are when they sign up, and use this to think about the groups which might form. If you have no sense of what groups may form--this is a good thing--you can judge your hardware and resource requirements purely on your aims and expected audience."

Of course, unlike software hackathons, which only require a laptop and Internet connection, the parts and tools required at a hardware hackathon could be endless.

"Making sure every hacker brings, can get, or is supplied with the parts they need to build their hack is a huge issue," Homuth says. "We've had the most success encouraging hackers to bring parts with them instead of supplying them, and then having a 'store' at the hackathon in case they forgot or realize there is something else they need."

But Homuth points out that parts are just one part of the hardware puzzle.

"Parts are fabricated, connected, and tested, with tools," Homuth says. "Again making sure every hacker has the tools they need is hard. Especially if the hackers don't yet know what they are building when they show up."

Homuth says that if you know your aims, and budget accordingly, most organizers will not have a problem making sure hackers have the tools they need. However, if your budget is strained, Homuth notes that you may need to alter your plans if you need some of the biggest, most expensive, and hardest to provide test equipment, like scopes, and fabrication tools, like milling machines. He also says to avoid 3-D printers ("too slow."), PCB board etching ("too messy"), and welders ("too heavy duty").

Above all, Homuth says, "Aggressively communicate what the hackers need to bring."

4. Work Out Lighting, Music, Food, And Use Of Space.

"This is a high priority," says McWilliams. "Really great things happen when people feel creative, and feel like they are in a creative community. This is about crafting a mood and context in which creativity can flourish. It should feel more like a social event than a workplace, and humor is actively encouraged."

For this step, start with the space. Find a venue that is big enough to hold all your hackers and the equipment they'll be using. Remember, these aren't just laptops they'll have. They'll need room to solder and spread their parts out. Creative people generally move around a lot when they're thinking. Make sure people have room to walk. Lighting and layout add a lot to making people feel comfortable, which breeds creativity.

But don't forget about practical logistics, McWilliams says.

"You may need good airflow for clearing the room of soldering fumes, and cleaning products or assembly tools. People will need a good amount of desk space to lay out their gear, prepare, and get going. It's hard to hack on anything where things around us feel precious, or when surrounded by distractions. We need cleared surfaces to tinker and solder, but also a space with a maker vibe so we can feel free to be messy and break things apart," he says.

"Make it easy, and more than that make it fun. A hackathon should feel like a social event. We aren't focusing narrowly on product delivery here, what we want to do is set a mood and context in which creativity can flourish. Our hackathons are about experimentation--play, not work. This means careful and subtle use of music and lighting, and leading by example with encouraging, friendly, constructive interactions. The atmosphere should feel distinctly different from a daily workplace, even if that workplace is generally a nice and comfortable space," he adds.

5. Figure Out Your Starter Pack.

"This is both for late-comers, and to remind people where to find things after the initial orientation," McWilliams says. "People have limited time to make something innovative, don't make them search for resources! This can be info on the wall, USB sticks, links to source code, APIs, explanatory blog posts, et cetera."

Steps one to five are where the meat of your planning should go if you want to throw a killer hardware hackathon. If you've done those well, now comes the fun part, when you get to see the fruits of your labor.

6. A Little Orientation Doesn't Hurt.

When hackers arrive to where they'll be working it never hurts to orient them to the space and everything that is at their disposal. It's also a great time to reiterate your broad mission and philosophy.

"Do a high-level orientation on the technology," McWilliams says. "Have the hardware already out on the table and demos ready to go. Try to do all of this as quickly and concisely as possible--each group can explore details ad hoc. Keep the energy up!"

7. Form Groups Around Participants' Motivations.

"Each person is there for a reason and you will get the most out of them if you can discover that reason," McWilliams says. "Where possible, ask people to express their motivations to the group so you are best-placed to connect participants with adjacent interests. Allow groups emerge from that, and then ask each group which aims they would like to explore."

8. Now Get Out Of The Way And Let Them Hack.

Once everyone is settled in, now it's your time to become the janitor Homuth stresses you must become. Creative people working toward novel solutions don't find those solutions by being given orders from overbearing organizers.

As McWilliams says, "To get them going offer guidance first, direction if needed, examples if requested, and orders never."

Up until this step meticulous planning was key. Now you need to let go, and watch the magic happen. As the magic happens you'll be playing janitor, waiter, and whatever else your hackers need. Keep the food and drinks coming. Help with finding parts if one of the teams need something. Mini-showcases and lightning demos are also a great way to keep people motivated, but never call time out on hacks. Constantly facilitate the flow.

As Homuth says, "Do everything you possibly can for them. And be incredibly grateful they showed up."

9. The Final Showcase.

While a final showcase is always part of the fun of a hackathon, remember to never put a strict deadline on it, McWilliams says. And instead of a showcase of "the best" it's more helpful to hackers to have it be an analysis of time spent. As McWilliams notes, "Outcomes which explore a possibility and find a dead end are highly valuable and not time wasted."

This is also the time to document, if you haven't already, just how inspired your hackers' works have been. Take tons of photos and videos, which will be important for the last step.

10. Share Your Hackathon Experience.

A hackathon is a lot of work from everyone involved, but the best parts of it don't happen only at the event. After the hackathon is over take all those photos and videos you took and post them online along with detailed breakdowns of everything your hackathon was about and all your hackers accomplished.

Actively contact all the event participants to share your photos, videos, and thoughts. This helps keep that sense of community alive long after the hackathon is over and will encourage participants to take part in future hackathons you organize.

Hacking House Plants To Make Music When They're Touched

0
0

All of our gadgets are glass, metal, and plastic. Wouldn't it be nicer to interact with a computing device that was alive and real?

After watching this video we went to see a talk by Ivan Poupyrev, former head of Disney Research, at Maker Faire in NYC.

When we got back, we whipped up four touch-sensitive plants that can be played like musical instruments. Real plants. Ones that need water and the sun for food and grow out of the dirt from little seeds. Our goal was to someday have people say things like "I'm mellow, I play the bamboo." Or, "I am hardcore so I play the cactus."

How The Musical Plant Works

The plants we hacked play different notes depending how they are grabbed. For example, when someone touches the stem the note is different than when they touch its outer leaves. When grabbed with two fingers instead of one, the sounds are different; same goes for other hand positions.

We reproduced a version of Disney's "Touche," the technology Poupyrev and his team built to encode the frequencies that conductive materials like water, human bodies, and plants, among other materials carry whenever they are touched by a human. The process was much like the one that allows your smartphone to be touch sensitive.

We build a Touche from a tutorial by Mads HoBye, Instructables' artist-in-residence, who hacked his own version using a small Arduino.

A Breakdown Of Touche With Arduino

Before jumping into the gritty detail of our project, we break down the code from the Instructable you will use to build a Touche shield.

The "Touche Arduino" is made of three components--the circuit board that amplifies the current caused by human touch, the Arduino which encodes this current as raw data, and Processing, which receives the data and parses it as an array of 160 frequencies via serial connection. A serial message sends sequential data one bit at a time, over a communication channel like a USB port.

Going Wireless

We wanted to keep things organic and keep the plants as far from the computer as possible. Our goal was to use the Yún, a new Wi-Fi-enabled Linux board combined with Arduino, to allow the plants to speak to the computer wirelessly.

Due to several challenges, we had to use a standard Arduino Uno hooked up to an iMac instead of our original plans with the Arduino Yún. We will explain the golden knowledge we learned in the process of failing.

Let's Get Real Time: A Story Of Trial And Error

We explored different ways to have the Yun inside the plants talk to the computer over Wi-Fi. The sole idea of playing an instrument is that you generate sound in real time. When you pull a string on a guitar you immediately hear the chord out loud. We could not afford to have our office musicians wait longer than half a second before they heard the first note and the next after.

Trial 1: Access Pin Via REST API

The data that's generated when you touch your plant is available via REST API because the Arduino Yún supports an Internet connection. The size of each encoded frequency is 160, so we needed to make a rest call 160 times just to get the value of a single frequency.

You can read about making REST API calls to an Arduino Yún here.

Trial 2: JSON

Our second option was making the Arduino structure the data before it returned it. When the user touches the plant, the Arduino Yún has the ability to structure an array of size 160 that stacks the data from the selected analog pin which the user calls.

The client makes the call:

myArduinoYun.local/arduino/plants.json

And the array that is created from that is returned:

[{1:153},{2:186}, …. {159:78}]

Although this reduced our original 160 calls down to 1, there was still a three-second delay because the Arduino was internally structuring that data into a stacked array.

Trial 3. Bridge / {Key:Value}

The Bridge library is a great predefined function on the Yún which allows communication between Arduino and Linux.

By invoking the Bridge.put method in your code, the data is saved to the Arduino captures as a 160 key-value pair array in the internal storage of the Linux. Then you can simply access each value by calling: http://myArduinoYun.local/data/get/KEY

And the Yún, or server side, returns the value inside the key slot in an array:

For example, invoking the call: http://myArduinoYun.local/data/get/159 would return 78.

This was still not good enough. There was a delay in the auditory response averaging 1.6 seconds and 5 in its worst-case scenario.

Trial 4: Shell Command

Since the Linux side of the Yún supports the shell command, this allowed us to write a lump of data to an micro SD card inside the Arduino.

We used a single line of code:

p.runShellCommand(echo "data" >> /mnt/sda1/arduino/www/logger/plant.json).

This was by far the longest running time we encountered, taking from 10 to 27 seconds before we got auditory response.

What We Learned

Although we did not end up using the Arduino Yún, we learned it is a great tool with many capabilities. It can be fast but not optimal for a project that requires real-time interaction. The connection will always be limited to the speed of the network. It did not help that our project dealt with such large amounts of data for each individual reaction to a frequency.

We are still hatching out a way to incorporate the Yún into our project because we believe there is still a lot of room for exploration with its capabilities and will write an update soon. In the meantime, here are the instructions to re-create our project with an Arduino Uno and a computer.

What You Will Need

Circuit board from Mads Hobye

Computer

Speakers

Copper Wire

Arduino

Processing

Minim Library for Processing

A Windowfarms stand for the plants (optional)

Plants

1) Install Arduino, Processing, and the Minim Library onto your computer.

2) Download our code from the Co.Labs GitHub page, an adaptation of Mads Hobye's Arduino Touche.

3) Build the circuit board from the Arduino Touche Instructable following the instructions here. We looked around and found the best shopping choice for the components you'll need:

4) Plug the USB cable to the Arduino board and your computer to run the Arduino code you downloaded.

5) Open and run the Processing code from our GitHub.

6) The paper clip should be hooked to your copper wire attached to your circuit board. Place the paper clip inside the dirt of your plant.

7) Check to see what values your plant is sending to the Processing console window. If you are not getting any value, check your circuit board and make sure your paper clip is securely inside the dirt.

8) Optional: If you want to pick the sounds your plant makes when it is touched in different places, add your own MP3 files inside the data folder inside your Processing sketch. We got our sounds from this open source website.

9) Compose your own symphony with your new musical plant!

Can These Consumer Apps Crack The Education Space?

0
0

Google Apps. Evernote. Dropbox. YouTube. You know them as technology products, but to millions of students and teachers they are platforms at the core of learning. A growing number of consumer-facing tech products are finding accidental success among educators--and may end up revolutionizing school forever.

GitHub joined the pack last week, lowering subscription prices for schools in response to student and teacher interest.

"We want to help the next generation of developers build the future," GitHub's education liaison wrote on the company blog, adding that 70,000 students had previously signed up to benefit from the site's informal discount policy.

Beyond a discount, though, educators don't always know what to expect from a consumer product. Should they put their faith in mainstream technology companies that view schools as a form of low-cost customer acquisition in a social-good wrapper, or rely on education-specific startups with unproven track records? The answer, as it turns out, depends on the teacher.

Tapestry: Accidentally Making Writing More Fun

When Jana Trantow agreed to run storytelling app Tapestry, she expected to manage a consumer product. But as she and her team set about growing their community, they stumbled upon a fervent group of early adopters that caught them by surprise: teachers hungry for digital storytelling tools.

With Tapestry, Trantow says, "You can have a voice and really learn the timing of a story." Those features made the app appealing to educators hoping to make storytelling relevant to students more comfortable with Pinterest than poster board. "Students don't feel like they're doing an assignment. It's writing, but it's fun," she says.

Moreover, the ability to share those tap-able stories with a broader community has kept students engaged and writing even after the semester ends--an outcome that's the stuff of dreams for many an education entrepreneur.

"It's not just submitting something for a grade," Trantow says. "It's getting love from the community."

Tapestry has ruled out a full-on pivot toward education, but the team does plan to push for growth in schools and universities over the next six months. Like many consumer technologies, Tapestry discovered the strategic value of the education market--scale, high engagement, relative stability--by happy accident.

It's a pattern that's more common than you might expect.

How Evernote Became A Surprise Hit Among Educators

Evernote's story is typical of technology companies that find themselves getting friendly with school leaders. The company created Evernote for Schools, a marketing extension of Evernote Business, after seeing organic trends in school adoption. Two years later, education is the single largest market vertical served by Evernote Business, with teachers and students representing one in 10 users. (Many more education market users are individual customers--university students in particular are loyal fans.)

From the company's perspective, converting a user early in life is a clear win. "It's great to have people start using Evernote," says John McGeachie, head of Evernote Business. "Hopefully it's something that's useful to them forever."

So far Evernote has sidestepped the potential administrative challenges of working with schools, which range from students too young for school-issued email addresses to parents concerned about privacy and security. Overcoming those barriers is up to school leaders, not Evernote product developers. "We don't want to get in the middle of it," McGeachie says. "We know we have a product that's great for teams of people, large groups."

Contrast that approach with Three Ring, a startup that offers a product similar to Evernote and boasts 35,000 users. With Three Ring's mobile apps, teachers can capture and manage the artifacts of students' work, such as photos of projects or video of presentations.

"Focusing on education alone is an edge because the classroom is a pretty sensitive and complicated environment," says Michael Lindsay, cofounder and CEO of Three Ring. He believes that teachers prefer technology solutions that understand the nuanced differences between messaging a fellow teacher versus a student, for example.

From Theory To Classroom Practice

That choice is one that computer science teacher Eric Allatta knows well. Allatta is part of the instructional team at the Academy for Software Engineering, a new public high school in New York City where students of every background can master the skills and concepts required to build software. In theory, it's the ideal home for an offering like GitHub Education.

"We've never used GitHub in the classroom, though I would love to," Allatta says. "I'd love to have kids contributing to open source." But for now, while the school serves just freshmen and sophomores (it will add one grade per year), Allatta prefers Scratch, Snap, and Bootstrap, all visual programming tools run by accessible university teams.

"All of the programming environments we use are very small operations--we're communicating with the developers," he says.

At the same time, along with millions of teachers around the world, Allatta uses Google Apps, with custom scripts he helped author and a teacher dashboard created by New Zealand startup Hapara layered on top. While Google can't offer Allatta the access to developers that he values in his instructional tools, it can offer him an open, flexible platform and an evolving stack of third-party solutions that together facilitate workflow management in his classes.

Sleeper success Hapara, a lightweight bridge between class rosters and students' work, lacks the brand awareness of a Khan Academy but is arguably one of the most successful startups in education. The two ventures are more similar than you might think--branding aside, Khan Academy is essentially a content layer with recommendation algorithms built on top of YouTube.

"One of the bets that we've made is that education as a whole is going to move to platform solutions," says Jan Zawadzki, the founder of Hapara and an engineer by training. "For these horizontal industry players, the cost of opportunity--for education--it's just too high. The ROI doesn't work."

Like Tapestry, Hapara launched as a broadly commercial product. Product tweaks that Zawadzki implemented as a favor to a few teachers led to an avalanche of interest. "We were in half a dozen countries even before we had a website," he says. "Educators are very well connected." Zawadzki declines to share user data, but points to trends in Chromebook sales and leaps in Google Apps usage as proxies for Hapara's expanding footprint.

Remix To Ignition?

In what perhaps portends the future, Allatta supplements Google Apps and Hapara with scripts specific to his school's needs. Functions that would otherwise join a long list of technical requirements managed by a district-approved vendor--"Send an email alert to parents when students are tardy," for example--are suddenly projects that a teacher can quickly implement. Cloud Lab, an initiative run out of New Visions for Public Schools, has been partnering with Allatta and other teachers to build out a library of these scripts--a sort of educational IFTTT ("if this then that") for Google Apps.

"They take more work, they take more tinkering, but they're extremely flexible. We're teaching other teachers how to use them more and more," says Allatta, who has started running professional development on the scripts for his peers. "At our school there's definitely more excitement than intimidation."

Could that same spirit of hacker optimism take hold in other schools? Teachers, after all, are accustomed to solving problems without much in the way of help or money.

Like it or not, they may have to. Even for a well-positioned vertical play like Hapara, venture capital dollars are in short supply.

"We definitely look at layers on top of education apps somewhat skeptically," Michael Staton, partner at Learn Capital, says of his venture firm's approach. "It's difficult to conceive how they can create a meaningful company without getting sideswiped."

Eileen Rudden, cofounder of LearnLaunchX, a Boston-based education accelerator, says she has found startups with deep pedagogical underpinnings to be attractive, given the landscape. She mentions CueThink, a recent addition to her portfolio, which is building a peer-to-peer learning platform for math problem-solving. "It incorporates new thinking on pedagogy and new student-centered approaches, which are not available from a company that is customizing what it has for the education sector," she says.

Some teacher advocates might groan at the idea of putting even more burdens on educators' shoulders--last I checked, time for writing scripts is not accounted for in most union contracts. But if there's a silver lining to this story, it's that technology is moving in a direction that more faithfully mirrors the way that teachers and school leaders like to operate. A generation ago, teachers asserted their professional voices by photocopying and re-ordering the chapters of their textbooks. Today, they can mash up software and design systems that are comfortably at home in their classrooms. The learning curve may be steeper, but over an academic year the efficiencies are potentially worth that price.

Let's just hope that the platform players, with those heartwarming education landing pages, keep up their end of the bargain as a foundation for teachers' ingenuity.

Inside DuckDuckGo, Google's Tiniest, Fiercest Competitor

0
0

When Gabriel Weinberg launched a search engine in 2008, plenty of people thought he was insane. How could DuckDuckGo, a tiny, Philadelphia-based startup, go up against Google? One way, he wagered, was by respecting user privacy. Six years later, we're living in the post-Snowden era, and the idea doesn't seem so crazy.

In fact, DuckDuckGo is exploding.

Looking at a chart of DuckDuckGo's daily search queries, the milestones are obvious. A $3 million investment from Union Square Ventures in 2011. Just prior to that, a San Francisco billboard campaign. Inclusion in Time's 50 Best Websites of 2011. Each of these things moved the traffic needle for DuckDuckGo, but none of them came close to sparking anything like the massive spike in queries the company saw last July. That's when Edward Snowden first revealed the NSA's extensive digital surveillance program to the world. The little blue line on the chart hasn't stopped climbing north since.

"Every year, we've grown 200-500%," Weinberg says. "The numbers keep getting bigger." As of early February, DuckDuckGo was seeing more than 4 million search queries per day. One year ago, that number had just barely broken 1 million.

Surprisingly, the sudden success didn't send the site crashing down. Nor did it change the company's stripped-down, razor-sharp focus. Here's how one small company is slowly, surely beating its way into the most monopolized category in technology.

Three Ideas In One: Where DuckDuckGo Came From

Weinberg didn't originally set out to build a search engine. After shuttering one failed startup and selling another to Classmates.com for $10 million in 2006, the MIT grad found himself exploring several new ideas. Across multiple projects, he focused on structured data, Quora-style Q&A, and programmatically combating spam.

"I started all of these projects independently and none of them really took off," Weinberg says. "Then I realized, maybe if I put them all together, there might be an interesting search experience there."

The result was DuckDuckGo, a search engine offering direct answers to people's queries, rather than merely delivering a list of links. Below these so-called "Instant Answers," the site still displays traditional, link-by-link search results syndicated from third parties like Bing and Yandex but, crucially, they're filtered and reorganized to reduce spam.

Looking at the modern history of the search space, Weinberg noticed that several companies in the early 2000s had tried--and failed--to rival Google by mimicking its method of mass-indexing the web. Instead of following this ambitious (and very expensive) path, Weinberg decided to let other companies' infrastructure do most of the heavy lifting so that his startup--initially, just him coding at home with his newborn son nearby--could focus on building a superior experience for finding information online. The key, as he saw it, would be Instant Answers.

"When you do a search, you generally want an answer. You don't necessarily want to click around links," Weinberg says. "It's our job to try to get an answer. Our grand vision is that that happens for 80% of queries, even for very niche things."

To deliver these answers, DuckDuckGo relies on a mixture of third-party data sources and the deep--sometimes bizarrely arcane--knowledge of its growing community of users and developers.

"We have recipes and Lego parts and other weird stuff that we don't know about as a team," Weinberg says. "The only way that works is if we have a community of people who are interested enough to know about these subjects. And then come up with ideas about what the answer should be, suggest the sources, and even develop them."

With Instant Answers, he was clearly onto something. Today, when you run a Google search for "Galileo" or "How tall is Big Ben?" the usual list of blue links is accompanied by a tidy white box displaying a mini-bio of Galileo or, in the latter case, the answer: 316 feet.

This is the biggest risk for any startup who dares challenge a giant head-on: At any point, the Googles or Facebooks or Apples of the world can just mimic what made you different, slam-dunking your shattered dreams into the wastebin of tech history.

Weinberg and his small team seem undeterred. After all, DuckDuckGo has one asset that Google could never copy, even if it wanted to.

DuckDuckGo's Secret Weapon: Hardcore Privacy

When you do a search from DuckDuckGo's website or one of its mobile apps, it doesn't know who you are. There are no user accounts. Your IP address isn't logged by default. The site doesn't use search cookies to keep track of what you do over time or where else you go online. It doesn't save your search history. When you click on a link in DuckDuckGo's results, those websites won't see which search terms you used. The company even has its own Tor exit relay, allowing Tor users to search DuckDuckGo with less of a performance lag.

Simply put, they're hardcore about privacy.

But things didn't start out that way. Weinberg, who says he has "always been a privacy-minded person," wasn't particularly concerned with search privacy issues when he first started building the service. In fact, he knew very little about the matter at all. Then early users started asking questions.

"Some of the first questions I got were about privacy," Weinberg says. "I had launched and it was on Reddit and Hacker News. So it was tech-heavy and there were a lot of privacy-minded people."

Did the site use tracking cookies? Did it log IP addresses? These are things Weinberg hadn't thought much about, but clearly mattered to his newfound user base. So the self-described tech policy geek decided to dig deeper into the privacy practices of Google and other search engines. He didn't like what he learned.

"If you look at the logs of people's search sessions, they're the most personal thing on the Internet," Weinberg says. "Unlike Facebook, where you choose what to post, with search you're typing in medical and financial problems and all sorts of other things. You're not thinking about the privacy implications of your search history."

This common functionality, combined with the possibility for accidental data leaks and hacks, made Weinberg nervous. At the same time, he realized that it was still possible to build a viable business model around search without tracking users. One might not reach Google-levels of profitability without large-scale targeted advertising, but the fundamental logic of delivering search ads--user searches for motorcycles, you show them a motorcycle ad sold by keyword--is still sound. And, of course, the profitability of such a model only grows as more people search more.

It quickly became clear that taking a no-holds-barred approach to privacy would give DuckDuckGo a unique selling point as Google gobbled up more private user data. So the company positioned itself accordingly and started amassing attention as the issue of online privacy slowly ballooned in the public's consciousness.

"It was extreme at the time," Weinberg says. "And it still may be considered extreme by some people, but I think it's becoming less extreme nowadays. In the last year, it's become obvious why people don't want to be tracked."

How DuckDuckGo's Community Informs Its Product--And Payroll

One person who didn't think the pro-privacy tack was extreme is Caine Tighe. After hearing about DuckDuckGo in 2009, the young programmer decided to reach out to see if Weinberg needed any help.

"I met Gabe like I meet a lot of people: on the Internet," says Tighe. "I thought it was interesting that he was working on a new search engine. He didn't really have that much stuff for me at the time, as he was still fleshing out the idea. Later on, I helped him write the first Android app."

By 2011, armed with its first infusion of venture cash, DuckDuckGo was ready to leave Weinberg's basement and grow into a real company. Having worked on multiple projects with Weinberg, Tighe was an obvious choice to become the company's first full-time employee. He soon became the Director of Core Components, a role in which he oversees DuckDuckGo's infrastructure, from its underlying web technology to the functionality of its mobile apps.

Tighe's trajectory is one that all future employees would follow: One by one, members of the DuckDuckGo community have gradually increased their involvement with the site--building plug-ins, recommending obscure data sources, even contributing code--eventually positioning themselves as no-brainer job applicants the next time the company is hiring.

"If you want to work with us, we find that the most motivated people come directly from the community," says Zac Pappis, DuckDuckGo's Director of Marketing and Community. "Anybody can get their feet wet." Indeed, Pappis himself was heavily involved in the DuckDuckGo community and did part-time work for Weinberg in the evenings before being offered a full-time position in 2012.

It's in this capacity that Pappis helps oversee the growing community of users and developers that contribute to DuckDuckGo. Over time, this community has taken on an increasingly vital role in the evolution of the company's core product. Not only does it pluck new employees from this group of devotees, but a subgroup of it has splintered off into a platform all its own: DuckDuckHack.

Letting Users Hack Their Own Search Engine

"It started very casually," says Weinberg. "I just made some APIs and tried to get these guys involved coding stuff for it. And then it clicked that this could actually be a platform that people could work on."

Branded as an "open source DuckDuckGo," the DuckDuckHack platform lets developers build their own Instant Answer plug-ins using the company's suite of (mostly Perl-based) APIs and documentation. It's essentially a toolkit that lets anybody programmatically define how to interpret whatever people may ask a search engine, as well as where to find the information online.

Some examples are obvious: Information about movies comes from IMBD, or Rotten Tomatoes. A search for "The Beatles" pulls in a biographical blurb from Wikipedia or Last.fm. Simple computations come from Wolfram Alpha. Tighe seems particularly excited about their recent integration with Forecast.io, the data source behind the popular weather prediction app Dark Sky.

Then there's the more obscure, less obvious stuff. That's where the community comes into play.

"The Wolfram Alpha experience really opened my eyes to the fact that you could have answers for really esoteric stuff," says Weinberg. "And we started including them. You could have answers for tons and tons of topics that our team knew absolutely nothing about. Celebrities, for example. We're not big on pop culture here."

One user, for instance, was a big Lego nerd. As Lego enthusiasts know, each piece of the iconic building block set has a unique ID number. To make looking up those pieces easier on DuckDuckGo, he crafted a plug-in that pulls from a database of Lego pieces built by Lego fans every bit as geeky as he is.

Scrolling through the list of existing Instant Answer queries--known as "goodies" on DuckDuckHack--you begin to understand why DuckDuckGo has opened up this part of its process to others: Some of this stuff is just too weird for one company to tackle.

Inside DuckDuckGo: How Its Tiny Team Works

Stepping inside DuckDuckGo's headquarters in Paoli, Pennsylvania, it's hard to imagine that its two modest, sparsely occupied floors could be used to launch an insurgency against one of the biggest tech companies on the planet.

But they are--sort of. Most of DuckDuckGo's 20 or so employees work remotely. Doug Brown, a front-end dev specialist, lives in Toronto, making the trip to the small Philadelphia suburb periodically. The company's newest hire, a developer named Jag Talon, started working for the company from the Philippines before his family relocated to New Jersey last year. When we visited their office in January, fewer than a dozen people roamed the halls. This, we were told, was more than usual.

Day to day, the team's priorities are dictated by what they call their "critical path"--the short list of focal points that prevent them from getting distracted from DuckDuckGo's core mission. Surprisingly, their recent surge in traffic hasn't altered the critical path very much, other than to initially double-check that the site was well-architected enough to handle the load (It was).

One recently checked-off item from the priority list was the consolidation of the company's public-facing community sites into a single Web interface. Next up is a visual redesign of the DuckDuckGo website, which they're aiming to relaunch in the second quarter of this year.

Like any company with a mostly remote team, DuckDuckGo experiments with all the latest online collaboration tools.

Skype. Yammer. HipChat. Asana. "We've tried everything that we know of," says Pappis.

Lately, they've been toying with Sqwiggle, an online collaboration tool that uses persistent video and periodic screenshots to let coworkers see each other--or know who's away from their desk.

"Skype is more like picking up the phone, whereas Squiggle is more like texting," says Brown. "You can see people's faces and you can chat with them really quick without having to set up a formal time because you know they're right there. That's pretty much the key to everything. And making sure that everybody knows what you're doing and when you're doing it."

On the tech side, all of the team's code--mostly Perl and JavaScript--is managed and discussed through Git and GitHub Enterprise. Every change, no matter how minute, is pushed through the team's HipChat room so everybody can see what's going on at all times. This practice, everyone seems to agree, is incredibly handy.

"It's interesting because that's a by-product of the fact that the team is remote," says Tighe. "It wouldn't exist if everybody were in this office, because everybody would just be pinging each other about what was going on locally, as opposed to being like: We need to have every single piece of information that goes out available to everybody on the team that could potentially be responsible for it, asynchronously."

When it comes to the tools individual developers use to build out DuckDuckGo's features, Weinberg isn't picky.

"We try not to control people's environments," he says. "People are remote. We give everyone their own dev environment that's a full stack of DuckDuckGo. It's a full machine. Some people are using Emacs. Some people are connecting directly to the machine. Some people are developing locally. People work in all sorts of different ways."

Inside The Swedish Gaming Mafia That Produced Minecraft And Candy Crush Saga

0
0

Two of the world's most successful games--Candy Crush Saga (by King Games) and Minecraft (by Mojang)--were created in Stockholm by two very different companies. King Games just filed for an IPO which is expected to raise $500 million, and Mojang has gone to great lengths to remain independent with no investors apart from the founders.

So what do these two massively successful game makers have in common besides a postal code? We talked to the creators to find out.

Some Quick History

The first thing you need to know is that Mojang, the company that built Minecraft, was actually started by creator Markus "Notch" Persson after a four-and-a-half year stint as a game developer at King Games (or "The Kingdom," as employees call it.)

Yet with Mojang, Persson seems to have set out to create a games studio that's almost the exact opposite of King Games. But how can the antithesis of a hugely successful gaming company also be a hugely successful gaming company?

How King Games Works Internally

Mojang and King have at least one thing in common: Their success is based on a single monster hit. King makes casual, puzzle-style games mainly played on Facebook or mobile. Although it has a portfolio of 160 games, Candy Crush Saga, which is the number one game on Facebook and iOS, still generates up to 78% of the 10-year old King's revenue.

"The biggest play base is mobile," says Sebastian Knutsson, chief creative officer at King Games. "On mobile when you have 2-3 minutes you can't really play a deep strategy game. People want that bite-sized entertainment and we fit that."

Seventy-five percent of King's 100 million daily players gamers are women aged 25-55 and 50% have kids.

In Hong Kong alone, King has 1 million daily players out of a population of 7 million. We have almost touched every smartphone in those countries. "People ask strangers for extra lives in the subway," says Knutsson.

With casual games like Candy Crush, gamers often only play for less than three minutes at a time but it can still take six months to get through a complete saga game. Candy Crush alone has more than 400 levels.

"We have a fairly standardized process of making a saga game," says Thomas Hartwig, King's CTO and cofounder. "King.com acts a testbed to try out a game concept. For creating this first game it's a very small effort. The games which we see the audience like to play, we invest more heavily in them. Once we take the decision to make it a saga we take a two-minute game experience to a two- to six-month game experience."

At King, typical games team consists of eight to 10 people, including graphical artists, back-end devs, game devs, producers, and a data scientist who is added during the soft launch. The data scientist's job is to help balance the gameplay.

"Look where people are having difficulties," says Hartwig. "Looking at the specific levels if they are too hard to manage, looking at different funnels within the game, seeing that people understand the game. At some point we start looking at other stuff too--of course monetizing features--but the gameplay is extremely important for us. King also gathers more than 1 billion data points per day on its players and uses them to optimize engagement across its multiple games.

How Mojang Does It Differently

If King feels a bit like a highly efficient, game-production factory, the three year old Mojang is a studio built around Persson's personal mission to make the world-conquering Minecraft. Minecraft is an open games world where players build out of textured cubes as well as performing other activities like exploration, gathering resources, crafting, and combat. Persson built Minecraft as a side project while toiling as a web developer and eventually quit King to start Mojang in 2010.

Minecraft is now one of world's best-selling games with 14 million copies sold for PC/Mac and 35 million copies on other platforms like Xbox and iOS. It's particularly popular with children of all ages (and sometimes their bemused parents).

London's Victoria and Albert Museum just hosted an event celebrating all things Minecraft. South Park wrote an entire episode about the game. The United Nations uses Minecraft to involve young people in urban planning. Swedish politicians use it to solicit suggestions on how to improve some of Sweden's most troubled neighborhood.

Who Wins The Money Game?

Neither Knutsson nor Carl Manneh, CEO of Mojang, like to talk about the money, despite the fact that both companies are making piles of it. King refuses to give out exact numbers on revenue from particular games and how many people pay for them. Manneh says that "most of the people at Mojang are not very interested in the money aspect. For the outside world that's a stamp of success, but for us I think its most important to grow the community and have as many players as possible."

The Atlantic reported that King made $1.88 billion last year, $568 million of which was profit. According to Swedish tech news site IT24, Mojang made $240 million in revenue and pretax profits of $89 million in 2012, with Persson personally receiving $98 million (640 million kroner) in Minecraft licensing fees. Mojang's revenue comes entirely from sales. There are no free versions of the game.

King's games are mostly free. "Eighty percent of the people have reached the end of Candy Crush, and it's around 400 levels now, without ever paying anything," says Knutsson. "We try to balance the game so it shouldn't be a forced pay experience. It's more about retaining the users and keeping them active. The people who pay they usually didn't have the hours or they didn't have the patience to keep trying a level that was difficult." King no longer uses advertising because "an ad in a game is always a disruptive experience," says Knutsson.

The Ethos And Offices

Neither Persson nor his cofounder, Scrolls designer cofounder Jakob Porser, had any experience with running a company so he brought in his former boss Manneh as CEO. Mojang's holy trinity of Persson, Porser, and Manneh make all major decisions at the 35-man company. "We can change the strategy of the company in one evening," says Manneh. "That's the strength of Mojang. We are independent."

Persson seems happy to hand off the business side to Manneh and no longer speaks to the press, only to Minecraft's community. "Markus lets go of things that he knows that other people can do well," says Manneh. "He wanted to focus fully on the game. Markus already had a culture set with the game and the way he interacted with the community and being very transparent with everything we do, almost like our internal communication is external. Anyone can affect anything from the game to the way we manage processes in the company. It's easier to do that when you don't need to answer to investors or a publisher or the stock market."

Everything at Mojang from business decisions to decor is driven by the games and their developers. "One of our visions is to be the best company to work for in the world," says Manneh. "So far no one has quit." In Mojang's first year, Persson gave away his entire dividend of about 25 million kroner [$3.84 million] to the employees. To celebrate selling 10 million copies of Minecraft, the company hired seven private jets to fly the entire company and their partners to Monaco, where they partied on the the biggest yacht in the Riviera.

Minecraft's office looks like a cross between an English gentleman's club and the set of a 1960s James Bond film. Pairs of Chesterfield sofas sit on a tartan carpet beneath hand-painted oil portraits of each and every staff member. The company even has its own crest. There's movie theater/games room where the entire company plays games on Friday afternoons.

In contrast, King Games' Stockholm office feels much more like that of a corporation. You can't tell when you walk in that it's a games company. Twenty percent of its 200 employees are female, a figure that Mojang certainly doesn't match.

King is a much larger enterprise than Mojang with 550 employees and games studios in London, Barcelona, Bucharest, and Malmo. It has a complicated corporate structure with holding companies in Ireland and Malta and big-name venture capital investors like Index Ventures.

All the founders are well versed in the business side of the company. King's management team has been with high-growth companies before. "We did all the mistakes in the late '90s during the dotcom boom," says Knutsson, "so we are trying to not repeat them."

The Future

Apart from the IPO, King's focus for the next year is firmly on mobile and on Asia. "Right now mobile is the tech vector that drives change in the games industry," says Knutsson. "The markets we have had the best penetration in so far is Hong Kong and Taiwan." Half of the world's Android users are to be found in Korea and Japan alone.

Mojang launched Minecraft Realms, their own online multiplayer service, in beta at the end of last year. Currently it is only available in Sweden, with a global rollout planned in the near future.

To maintain their success, both companies will have to live up to the Mojang's Latin motto e pluribus ludum: not one huge success, but "one of many games."


Should You Major In Video Game Design?

0
0

Consumers spend over $20 billion on video games, so budding programmers are eager to dive into the field. Meanwhile, universities across the country, who offer degrees in video game design, are just as eager to scoop up applicants. But as the Internet and mobile platforms continue to disrupt the gaming landscape, these schools are being forced to adapt. Can they compete with the speed and edge of a self-taught experience?

Top Gaming Programs

"Mobile is definitely a major change," says Doug Schilling, a former Microsoft engineer. He's now a senior lecturer and the game design department chair at DigiPen Institute of Technology in Redmond, Washington.

Only 26 years old and with a student body of just over one thousand, the private DigiPen Institute started as a computer simulation and animation studio, and gradually started offering courses in 3-D computer animation. It worked with Nintendo of America to establish post-secondary study in video game programming in the early '90s. Today, it's one of the country's leading schools that specializes in interactive computer tech.

"Another example is the shift from a traditional 'full purchase' or subscription model, to a free-to-play model for casual games and massive multiplayer online games," Schilling says. ("Free-to-play," or F2P, refers to games that charge customers little to nothing. Angry Birds Go! is an example.) So DigiPen is integrating solutions to these modern challenges into their curriculum: "Students are learning why these monetization techniques are of such interest to publishers, and how game design and development convert an F2P player into a paying customer."

And major universities across the country--MIT, Michigan State University, North Carolina State University, the University of Wisconsin, the Savannah College of Art and Design, and scores of others--now offer both undergrad and graduate programs in video game design and production of their own. The Princeton Review even compiles a list of America's top gaming programs, and the updated list will be released next month.

Video games' seep into academia has been a steady drip over the last decade, so their existence is nothing new. But the video game industry has exploded into a market that can outperform motion pictures, and several widespread technological trends have disrupted the way people play, build, buy, and sell video games.

Competition From Kickstarter

For example, in today's world of crowdsourcing, veteran and rookie devs alike can make a game in their garage, launch a funding campaign on Kickstarter, and then smack the world in the face with the next Flappy Bird. When success can come so easily to those in an increasingly crowded and competitive arena, does an expensive university degree in video games really matter?

Schilling says yes, and likens a big indie break on Kickstarter to winning the lottery. Universities also point to the argument that a college degree of any type, especially in an idiosyncratic field like video game design, increases the likelihood of success by equipping students with the necessary skills and contacts.

"The relationships they make while here form tight bonds that will last a lifetime as they go into the industry together," says Tracy Fullerton, director of the Game Innovation Lab at the University of Southern California.

Doug Schilling points out that the games industry is "highly collaborative," and immersion in a team-driven environment arms students with the soft skills they'll need in the real world. Plus, they'll have that all-important portfolio to show employers a body of work from the get-go.

Another challenge is the fact that game design is such a specialized area of study, meaning it could be hard for kids to find a job in such a competitive field. That's why programs encourage a multidisciplinary approach.

"It's akin to going to a conservatory to study art or music," Fullerton says. "Students of game design are often also students of creative writing, art, psychology, technology, history, media theory, economics, and more."

Defining Success For A Gaming Education

The Princeton Review keeps the evolving nature of the video game industry in mind when dubbing 10 schools the best in the country for studying video games. These schools are "not only teaching students the core concepts of game design, but also how to write a business plan for that game, or how to develop a funding prototype," says David Soto, director of Content Development at the Review.

Good gaming academic programs introduce students to all the standard technologies: C++, Java, Maya, ActionScript, Unreal Engine, Flash, Unity, XNA, Torque, Processing, and Gamebryo. Applicants should also look for programs that have more recent tech at their disposal, like augmented reality devices (including mobile), Google Glass, Oculus Rift, Kinect, Leap Motion, and CAVE. The inclusion of all these in a school's program means the school has a grip on the industry's shift toward motion-capture gameplay and virtual reality.

Tracy Fullerton says some of the most popular classes in USC games' department are experimental game design and immersive game design, and points to wildly successful alumni like thatgamecompany's Jenova Chen, who's turned the industry upside down with games like the award-winning Flower and Journey. Those games' artsy presentation, Zen-like gameplay, and nature-driven mechanics snagged accolades and set a new creative precedent. Chen received an MFA in interactive media at USC in 2006.

"We don't teach technology for technology's sake," says Fullerton. "Games are an aesthetic form that can supercharge and elevate our human and social experiences, and that is reason we do what we do."

A couple of decades ago, the video game industry's economic and cultural pervasiveness wasn't what it is today, and video games as a topic of academic study certainly didn't exist, at least as it does now. What started as a niche has morphed into a dynamic field with ever-growing challenges. "It's nice to know that students today have a much more direct route into the industry," Schilling says.

If young programmers are thinking of getting a degree in game studies but are on the fence, the statistics are pretty convincing, if nothing else. The industry is changing, but having a degree in video game design gives you work samples and connections you wouldn't get otherwise.

"Of our top 10 undergraduate programs, 75% of their most recent graduates worked on a game that shipped before graduation," says Soto. "And 82% of those students at our top 10 schools had a job waiting for them at graduation."

The Search For Talented Writers In The Prison System Is Changing It, Too

0
0

Jack London, Ken Kesey, William Burroughs, Oscar Wilde, Malcolm X, Voltaire, Cervantes, E.E. Cummings, Martin Luther King, Mark Twain--all literary legends who spent time behind bars. Some only a few nights, others years. A few even penned classics while in the joint.

Unlike those lucky standouts, most incarcerated art doesn't make it outside the confines of a cell. But that may be changing. Random House recently paid six figures for the rights to a handwritten prison memoir titled "The Life and Adventures of a Haunted Convict, or the Inmate of a Gloomy Prison."

Now a Texas publishing company is searching for the next great incarcerated writers--something never possible before software, thanks to strict laws about communicating (and, of course, planning entrepreneurial creative efforts) with prisoners. Here's how they're pulling it off.

The Hunt For Creatives In Captivity

Houston-based Vidahlia Press and Publishing House and the digital publishing software company Pubsoft have teamed up to host INK, a writing contest open to any and all inmates in prisons across the country.

Winners of the Vidahlia contest will be featured in an anthology to be published in print and online, but also via Pubsoft, a web platform which is offering the new inmate authors presale, sponsorship, and marketing help--areas where even big traditional publishers can be flaky.

"It's a talent identification program," explains Roy Rodney, Vidahlia's founder, who says he is "blown away by the history of writers in prison."

Inmates are listening--and they've been writing and submitting in huge numbers. More than 5,000 incarcerated men and women have sent pieces of poetry, non-fiction, fiction, dramas, and graphic novels. Now it's up to Vidahlia to get to work, reading and picking winners, no easy task. Especially since most entrants didn't send neat printed documents or PDFs. Submissions were largely scribbled "by hand on loose-leaf parcel," says Rodney. Others "typed, sent via fax, email, and family members."

Winners will also receive cash prizes of up to $500 per category, and the option of free tuition for yearlong online or correspondence programs in fine arts or creative writing. Winners will be announced on July 4th.

Software Makes It Possible Where People Couldn't

The year-old publishing house combed through entries with the help of Houston University English majors, who spent Saturdays reading and screening before sending to Brenda Marie Osbey, the poet-in-residence at Brown University, and the editor of the anthology. The top 25 submissions were then run through Pubsoft's plagiarism detection software to make sure they were kosher.

"As a startup press we wanted to do some things differently in reaching out to the artistic community," Rodney says in a thick Texas drawl.

It's different, all right--and borderline illegal. That's because prisons are very strict about incoming and outgoing communication, and because of restitutions and state legislature, making sure inmates don't profit off of their crimes is a judicial tightrope. But that didn't stop the dedicated people behind INK.

"It's better to ask for forgiveness than ask permission," says Rodney, whose small staff reached out to 2,200 prisons, jails, halfway houses, unions, civil rights associations, "anybody traditionally a stakeholder or an advocate for or against the state of the prison system."

They took out ads in prison newsletters and periodicals like Prison World and Prison Legal News, smuggling in contest literature with the help of the American Correctional Association, which gave mailing lists and posters advertising the contest to each institution's school or library.

"It's a way for us to shine a light on the kind of talent actually institutionalized," explains Rodney, who says Vidahlia plans on publishing a separate book of poetry, short fiction, and essays in addition to the anthology.

Instead of getting a slap on the wrist, in an unexpected twist, many prisons, and the men and women running them, have opened their arms to Vidahlia and INK.

"There is a great deal of talent in the system," says Daryl Greenridge, a staffer at the American Correctional Association. Greenridge says he fell in love with the idea of "turning negative energy into something positive," and became a valuable advocate for INK, lobbying for the program at prisons around the country.

"Writing naturally becomes a therapy," says Nick Lindsay, a PhD student studying English literature in Buffalo, New York. A few weeks ago he finished a pilot creative writing workshop at Erie County Correctional in Alden, New York. His students--prisoners.

Lindsay says that experience, while trying, offered a cathartic emotional outlet for the inmates. Daily writing instilled "a change in my students," he says. "There's a way that writing serves to prepare people to give concrete shape to the mind and soul. This way we can work through things that are otherwise unintelligible."

Uncovering Talent Behind The Wall

"We're the largest prison writing contest in the U.S., which might make us the largest everywhere," says Rodney. Next year he plans to make INK even bigger by incorporating Spanish-language submissions.

Dougal Cameron, Pubsoft's COO, says he'd like to expand the online reach of the contest, allowing Vidahlia to accept digital submissions ongoing. Cameron wants to "develop talent from the incarcerated population by providing an e-commerce space for the general public, to easily find and purchase digital versions of the works of incarcerated individuals."

In 2011, a St. Martin's Press subsidiary called Minotaur Books held a private-eye novel writing contest with a $10,000 advance prize and a guaranteed publishing contract. The winner, Alaric Hunt, was a 44-year-old literary phenom, and unbeknownst to Minotaur, a prisoner in a maximum-security facility in Bishopville, South Carolina. He was serving a life sentence for murder, but he was also a damn good writer. Good enough that Minotaur's editors overlooked his past. Because South Carolina allows prisoners to profit from their work and has repealed its own Son of Sam law, there were no legal complications. (That said, Hunt didn't write about his own crime.)

Still, exchanging edits and communicating with Hunt proved challenging, and personal appearances and book signings were out the window, a problem winners of INK will face. But three years later, Hunt's book "Cuts Through Bone," was released and is now on sale on Amazon, paving the way for other authors like him.

In a strange irony, Laura Pepper Wu, an editor of The Write Life Magazine, says the solitude of being locked up with nothing but your work could be conducive to writing.

"Stepping out of one's everyday reality and into confinement brings a quietness that allows for true reflection and insight, the pillars of good writing," she says, while also offering prisoners a boost in "positive esteem, identity, and self-worth."

Sending Kites

About 2 million people are locked up in state, federal, and private prisons in the U.S., about 25% of the world's prison population. "More prisoners than any democracy in the history of the world," laments Rodney.

While some are guilty of heinous crimes, and deserve punishment, many serving time are nonviolent offenders, addicts, and the mentally ill. Instead of providing rehabilitation, incarceration usually becomes a stepping stone to more crime. Most businesses won't hire felons, and the rate at which parolees return to lockup mirrors a revolving door, feeding the deep pockets of institutions more than helping fix offenders.

If inmates do make it to the outside, lacking confidence, job, and social skills, they often turn to the option that probably landed them inside: crime. One way to stop this pattern is to give the incarcerated desperately needed positive alternatives while they're imprisoned. Like the arts.

Rodney knows this cycle firsthand. Before he went to law school, became an attorney, and started Vidahlia, he fell into the same snake pit that has disproportionately captured so many young black men.

"It also happened to me," Rodney recalls. "A judge sentenced me for four months for filing taxes three months late. Instead of a federal compound, I went to a highly secured facility."

Rodney says he had restricted privileges, but was able to work at the prison library, which was in a "terrible state," but provided him a "therapeutic" escape.

When he got out, Rodney was determined to change the pattern and raise awareness, especially the rate at which institutionalization affects African-American and Latino communities. "Children of victims, fathers in prison, destitution in family, general malaise, hopelessness. My experiences have allowed me to be involved in something that can help others. We want to prevent people from going back."

An Inside-Outside View Of Incarcerated Artistry

Trevor Streeter was 19 when he was sent to Ironwood State Prison in Blythe, California in 2005. For Streeter, then an active gangbanger from South Central Los Angeles, it was his second strike. He was looking at five years for attempted murder of a rival gang member. No weapon was found in the case, and no witnesses testified, but because of his rep, the District Attorney pursued the case, sending Streeter on an up-north trip. It was his second felony.

As part of California's infamous three strikes law, another conviction would have put him away for life. He was still a teenager, but he was aware of how deep the water he was playing in. He says he decided then that he wanted to change his life. Barely an adult, he knew he didn't want to spend the rest of his days in a cell. But he admits he didn't have a clue how to get out of the gang culture that had swallowed his family for generations.

Streeter ended up serving a little more than two years at Ironwood before being paroled. He hasn't been back since. Streeter says what got him through those tough times was his secret: writing.

"A lot of the way my writing process developed was in jail," he says. "I never did any before I was incarcerated."

Today, almost a decade later, and his 28th birthday fast approaching, Streeter barely resembles that troubled teen. He's held down a job at UPS for the last year and a half, preloading trucks on the graveyard shift. And he's left gangbanging behind him. An aspiring rapper and writer, Streeter says when he was in jail, he wrote poems for other inmates to send to their girlfriends in exchange for soup, snacks, and candy bars. What really changed him was the self-reflection gained when he put pen to paper, alone in his cell, with nothing but time and his sins to keep him company.

"The only confidence I had left as a human being was what I was writing," he says.

Streeter, who says he uses rap as an emotional release, wished there was a program like INK when he was locked up.

"It's a good outlet," he says. "It can definitely boost their confidence, for one. Going to jail is not a confidence booster. Being able to come from a place of nothing with something makes prisoners feel like they can actually do something with themselves on the outside."

Inside Erlang, The Rare Programming Language Behind WhatsApp's Success

0
0

How do you support 450 million users with only 32 engineers? For WhatsApp, acquired earlier this week by Facebook, the answer is Erlang, a programming language developed in the '80s that is finally having its moment in the spotlight.

But with other languages starting to co-opt its ideas, does Erlang have a future?

Erlang was developed by Swedish telecom giant Ericsson over 25 years ago, and now it's finding a home at messaging apps like WhatsApp and TigerText. Even Facebook was singing the language's praises when it used Erlang to launch Facebook Chat back in 2009--the same year it turned down the job application of WhatsApp cofounder Brian Acton.

Concurrency Is The New Black

"With Erlang you could build a messaging app with many connected users and not worry about how they communicate," says Chad DePue, CTO of Whisper, an app for anonymous posts that has opted to use the language. Instead, he says, "You can worry about creating a great app."

Ericsson engineer Joe Armstrong developed Erlang with the logic of telecommunications in mind: millions of parallel conversations happening at the same time, with almost zero tolerance for downtime. Other programming languages can only give the illusion of managing all those conversations--some have gotten very good at keeping up appearances, but they aren't natural "multi-taskers." Erlang, in contrast, loves to multithread or "juggle" in this way--got a another spinning plate? Toss it over!

"The language is very expressive," says Igor Clark, a creative technologist. "You can talk at a high level and do quite a lot with its few key concepts."

In practical terms, that initially made Erlang very good at efficiently executing commands across processors within a single machine. Fast-forward to 2014, and it's making Erlang very good at executing an even more bewildering volume of commands across the global networks of servers we have come to know as "the cloud." For use cases in gaming, financial services, and anything that mimics the behavior of a real-time auction, that speed and reliability, at massive scale, is absolutely essential.

Equally attractive to engineering teams, and highly unusual, is the way that Erlang allows for bug fixes and updates without downtime. In effect, you can change the lighting on your set or execute a costume change without a pause in the action. This Erlang property is a legacy of the telecom imperative: As DePue says, "When you're on the phone, you can't have someone hang up on you because they're upgrading the system."

Turning The Tables On Telecom

Today the industry that open-sourced Erlang in 1998 is feeling less than generous. As WhatsApp, WeChat, Line, and other startups have blossomed, the messaging fees that telecoms once relied on have been rapidly eroding. KPN, a Dutch telecom, tried to block the upstarts from its network after seeing a drop in SMS revenue, but lost a battle in court over the practice.

Some established players, such as Vodafone, have tried to build messaging services of their own, with little success. Nearly all have been tinkering with the mix and pricing of SMS and data in their users' contracts.

While the telecoms were busying themselves with payment plans, WhatsApp was focused on product. Its engineering team has repeatedly wowed the Erlang community with feats of scale and speed, jumping from 10,000 to over 2 million connections per server in just a few short years.

Those achievements rest on the shoulders of Armstrong, whose signature project for Ericsson, AXD301, achieved a "nine nines" reliability (99.9999999%). But for now, at least, the young challengers are keeping the aging telecoms on their heels.

To Erlang, And Beyond?

With such a strong track record, you might think that Erlang is poised for breakout adoption. The reality is a bit more complicated.

Until now, the developer community around Erlang has remained small, and largely concentrated in Europe. The language's syntax is admittedly "weird." And if you need to analyze complex data or prototype a simple web app, there are better tools. Moreover, newer languages and variations have been borrowing from Erlang principles, leading to an increasingly cluttered landscape, from Google's Go to Docker.

There's hope for Erlang in Elixir--Armstrong recently bestowed his blessing--which normalizes the Erlang syntax and could help the language take off in the same way that Rails did for Ruby. Projects like Chicago Boss are also aiming to make the language more user-friendly. If a broader community does coalesce around Erlang, they will discover OTP (open telecom platform) libraries that more than make up for the language's limited presence on GitHub.

"They abstracted patterns from the software they were writing and made bulletproof libraries that you can piggyback on," Clark says. "It's the glue between the small parts and the top level."

The tipping point for adoption could be the one ecosystem with enough scale to dwarf mobile as we know it: the internet of things. Smart object hobbyists have been experimenting with Erlang as part of MQTT, a lightweight protocol for transferring messages between sensors. If you can SMS your thermostat, surely you can WhatsApp it, too?

The British Are Streaming . . . For Free! So Why Isn't NBC?

0
0

In the U.S., NBC has a death grip on Olympics coverage, aggressively keeping clips off YouTube and monetizing its broadcast at every possible turn. But the BBC tells a much different story. This winter, the BBC is broadcasting the Olympics live and in their entirety at no additional cost to viewers. To stream NBC's live Olympic coverage stateside, however, customers must absorb the cost of both a pay-TV subscription and Internet connection respectively (not to mention withstand commercials) for access to the online broadcast. Sports enthusiasts within the U.K., however, can enjoy not only their national BBC television broadcast, but each and every Olympic event streamed live to their preferred desktop or mobile device by purchasing a standard yearly BBC license (for only £145.50 per color TV per year, no less)--ad free.

Depending on their device and Internet connection, users can access the best-quality video available to them. The service is available to all mobile and desktop platforms (iOS, Android, Kindle), with streams' bit rate range from 56kbps (3G) to 3.5mbps for a 720p HD experience. "We have an offering of six live HD streams of every event in the Games, with 750 hours of live action. All streams will be delivered in RTMP, Adobe HDS, and HLS formats using Elemental encoders and distributed via multiple CDNs," says Neil Hall, executive product manager for BBC Sport. "We're now at that point where the technology and the devices that people have in their hands every day really can make viewing the content digitally a good experience, which really gives us the potential to delight audiences of each sport in a way that isn't limited by a network schedule."

With over 600 hours of streaming content to supplement the network's standard 200 or so hours of television content, the BBC is making good on its promise to provide access to the Olympic experience from anywhere, with 600 additional hours of live coverage--that's every event. Since their home Games in 2012, the BBC has experienced a staggering increase in viewership across the board.

"Cumulative TV reach so far is over 24 million, RedButton--our support interactive service--has reached 8 ½ million, and the website has hit 10 million," explains Philip Bernie, head of TV Sport for the BBC. "We have an appreciation index where anything over 80 is very high, and it's been regularly in the mid-80s. BBC 2, for instance, is beating by some distance the average [viewership] that the channel usually receives throughout the day. The benefit of the RedButton service is that we can, pretty much, show everything in different guises and so even the aficionado of, say, biathlon, which is not got a huge British following, will have the opportunity to watch that too."

And they're not done yet. The BBC has broadcast rights for both summer and Olympic Games through 2020, with hopes of extending well beyond that. "The BBC has shown the Olympics since they've been on TV," says Bernie. "We believe that the Olympic Games is a key part of what British sport is about and a time when the nation really unites behind its sporting events." And they do. Unlike American NBC's streaming coverage which requires a pay TV subscription in addition to Internet access, Bernie declares, "the key prerogative for BBC sport is to make sure that, where possible, we show to the nation--free to air and promotion free- the better sporting events that bring the nation together and really unite them."

An Open Source Library For Turning Journalism Into Podcasts

0
0

When Texas Senator Wendy Davis gave her 13-hour filibuster against an anti-abortion law last summer, journalists filled the web with a heap of multimedia: videos, tweets, GIFs, and Internet memes that often outshined their written articles. But the Washington Post blog "The Fix" didn't make its readers choose between listening and reading its filibuster coverage. By using SoundCite, a web tool that streams an audio clip behind the text, the Post's account of Davis's speech uniquely captured the event's emotions in digestible snippets without awkwardly breaking up its written text.

"When the Wendy Davis filibuster went down--which was inherently such an audio-rich event--we realized we could do something compelling with it," Caitlin Dewey, the blog post's author, wrote in an email.

The tool's original developer is Medill School of Journalism senior Tyler Fisher, who originally conceived SoundCite to improve the way online music articles weaved sound clips and their text together. An arrow appears next to the words that are animated by the audio, and a moving progress bar slides across the text, so you stay focused on reading. At just over a year old, SoundCite has not only made its appearance in the music-sphere, like at WBEZ Chicago Public Radio, but has also entered other types of newsrooms, like the Washington Post, JSOnline, and Education Week.

Startups aimed at content publishers are growing in numbers, with tools like Gui.de and SpokenLayer and Readrboard offering to give publishers easy access to the kind of functionality that users have come to expect from other parts of the social web.

Unlike these companies, however, Fisher doesn't aspire to make money off of SoundCite. In fact, he sees the project as core to a new type of storytelling. "A lot of people don't think about the usability of [multimedia], and I don't think they think about how the multimedia is contributing to the story," he says.

An App For Music Critics

SoundCite got its start when Fisher took an advanced design class at Medill in the spring of 2012. The class required that he develop a web tool to solve a journalistic problem.

"In a past life, Tyler aspired to be a rock critic," says Jeremy Gilbert, Tyler's professor for the class. "He spent equal time writing what the music sounded like and writing what the music meant. With the clips, he wanted to concentrate on what the music meant and let the music play for you," Gilbert says.

"A lot of sites have good players, but you usually have to play the whole song. That break in the narrative isn't good for the experience," Gilbert adds.

Fisher's earliest version of SoundCite consisted of JavaScript, HTML, and CSS that relied on the user to create MP3s loaded directly in the HTML page. "It wasn't powered by anything," Gilbert says. "Users couldn't easily pick out the start and end points they wanted."

"I was convinced that it could be used for much more than just music," Gilbert says. Gilbert decided to work independently with Fisher after the class ended. Eventually, they were able to connect with SoundCloud's API, making it easier for non-coders to create and choose time points within the sound file. What's more is users no longer had to wrangle MP3s onto their site; they could use SoundCloud's uploading feature and let them deal with hosting the clip.

Collaborating With Expert Coders

At that point, SoundCite still had JavaScript code that could interfere with other pages on a given website. Depending on where the tool was used, SoundCite's version of JQuery was likely to interfere with pages that didn't use that exact JQuery version.

Knowing that they needed help from the experts, Fisher and Gilbert applied to have Medill's Knight Lab host and support continuing development of SoundCite. "We needed to harden the code," says Gilbert. With no break in momentum, they transferred SoundCite from Gilbert's personal server to the Knight Lab that summer.

The Knight Lab, which is jointly funded by the Knight and Robert R. McCormick foundations, selectively accepts journalism-driven tech projects, whether they are from a Medill class or elsewhere in the professional world, and helps push them to usability. It employs two full-time software developers.

At the same time the Knight developers got up to speed on SoundCite, Fisher was accepted as a Knight Fellow for the 2012-2013 academic year. He essentially got paid to keep improving SoundCite while working in the lab.

Solving the JQuery issue meant isolating the version of JQuery that SoundCite used. "We had to make sure it kept itself tidy and didn't step on other things," says Joe Germuska, the Knight Lab's technology director, stressing that SoundCite's loose JavaScript could have interfered with news sites' advertising elements.

Germuska looked to the Internet for help and found a solution on Stack Overflow. "As is often the case, someone has already solved the problem. You just have to find it," says Germuska.

Another issue was rewriting the code so that non-coders wouldn't be scared off by it when it came time to paste it into their pages. "We really had to work on usability and create clean code," says Gilbert. "The Knight Lab really helped us with that."

Working On Mobile

"The main problem with SoundCite [now] is that it doesn't work with mobile," Fisher says. Mobile browsers don't support Flash, and the SoundCloud API uses Flash to pass its clips to third-party applications. The solution would be to create in-house code so that SoundCite wouldn't have to rely on Flash. This would mean hosting users' sound clips on internal servers, and this isn't an attractive option to Fisher.

"We definitely don't want to get in the business of hosting audio because of copyrights and all those legality issues," says Fisher. "The problem with audio codecs is that the MP3 is owned by someone," he says, "and some web browsers like Mozilla Firefox don't want to pay for that, so they don't support the MP3. Some browsers do play them. Making things more segmented: Some web browsers don't support the WAV file, so nobody agrees on what types of audio files should be supported." So it's good that SoundCloud uses Flash because Flash is universal, he says.

Besides, Fisher says optimizing SoundCite for mobile is outside of his scope. He is willing to give users documentation on how to host audio on their sites and has placed SoundCite's code on GitHub. "Everything the Knight Lab does is open source. So anyone can modify the code if they want to," he says. And someone has. Today, there is one WordPress plug-in for SoundCite on GitHub.

"The way people look at it is that the audio isn't integral to the piece, so it doesn't ruin the piece for mobile [when the clip isn't there]. But it does add something," Fisher says, bringing up the Post article.

With the code available to everyone, users can tweak SoundCite to work for whatever needs they have, whether it's mobile or fitting in with their web page's scheme. "Education Week was able to customize the look of SoundCite because its HTML was publicly available," says Fisher.

Keeping It Open

The Knight Lab plans to continue encouraging journalists to use SoundCite and promote collaboration, as they have done in the past. "We approached [WBEZ] directly and guided them through an early implementation," says Ryan Graff, communications director at the Knight Lab. "After that, the other projects were word of mouth." In the case of the Post article, a Post intern knew Fisher and mentioned SoundCite to her colleagues.

At the moment, Fisher is spending a chunk of the school year interning at National Public Radio and will return to Northwestern this spring. After graduating this June, he would ideally "be working with technology and news, somewhere in that intersection." And Gilbert has since moved on from Northwestern and is now deputy director of digital for National Geographic.

"I would hope that anyone who uses SoundCite would think first about what that is adding to their story and if that is necessary," says Fisher. For web journalists who want their well-picked words read, adding multimedia might keep readers on their page for a few seconds, but it does not guarantee that their prose will command their attention. But SoundCite does writers justice by illuminating the text, keeping the reader's focus on the writing. And that's what matters.

Viewing all 36575 articles
Browse latest View live




Latest Images