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

Redefining The Digital Music Service Starts Now

$
0
0

To engage enough fans to make money, music artists today are forced to spread their work across several different services, which means they lack unified data about interested listeners. But Spotify's new partnerships with Topspin, Next Big Sound, and Songkick could tilt the music industry--both publishing and distribution--away from the record companies and other music services.

With a unified analytics platform, Spotify hopes to offer three things to artists. First, it's a way to see who their listeners are and possibly how they became listeners--that's what they gain from the partnership with music analytics and data company Next Big Sound. Secondly, the new service will allow artists to go beyond selling just music and offer other items like concert merchandise for sale through the web using Topspin, a company that helps connect artists to their fans. Lastly, via the Songkick partnership, come direct ticket sales. It's exactly the type of thing we recommended Spotify do to placate artists after Thom Yorke's anti-streaming outburst this summer.

By comparison, iTunes is still a black box for artists. Today, an independent musician has to sign up for a third-party service to get music into the iTunes store, and then if anyone buys the music, they only get a simple payout report--no data, no segmentation, nothing. Apple helped kill the old distribution networks for independent music, but it hasn't yet figured out how to take these independent artists to the next level by empowering them as small business people.

Spotify is not without competitors, however. Beats' streaming music service will reportedly launch in early 2014 and with Trent Reznor helping to craft the service, so you can bet there will be plenty of focus on treating artists fairly. YouTube is already the place to find free music, but rumors and hints indicate the service will officially join the streaming music game soon as well. There's also existing services like Deezer, which will be coming to the U.S. in 2014, and Rdio, which cut staff and brought in a new CEO in an attempt to refocus. All the movement suggests this is a turning point for digital music and its relationship with those actually creating the music, rather than any rights holders removed from a musician's hardships.

Part of this transition also means there needs to be more transparency. Spotify has revealed how its payout system works, something it had kept vague up until now. Based on the information and charts provided it does appear Spotify is competitive against other new models like YouTube's ads, but streaming still isn't set up to create a middle-class music economy for independent musicians. This has caused a lot of artists to seek out music licensing which can help add additional funds beyond direct sales.

The frustration that artists feel toward iTunes is similar to what other creatives--namely, developers--have with the App Store. The lack of information provided to sellers is by far one of the biggest complaints with Apple's digital stores. No one knows how people found their content, what they searched for, or even where it's being featured, unless you find it on your own. Not knowing some of this trivial data hinders growth and being able to fully re-evaluate the direction of your app or album. If Apple doesn't make some changes to its user data policy, the huge number of songs in its music catalog will not be enough to keep musicians and other content publishers around.


Saturn's North Pole Has "Hexagonal" Storms That Look Like This

$
0
0

Unlike anything else in the solar system, Saturn's north pole is home to a six-sided jet stream in its atmosphere known to scientists simply as "the hexagon." First spotted by the Voyager mission in 1981, the hexagon has fascinated scientists but eluded attempts to capture it in detail on camera. Launched in October of 1997 and arriving in Saturn's orbit in July of 2004, the Cassini spacecraft was sent to study Saturn and its moons. And though it had brought us closer than ever, lighting conditions had kept it from taking a full, detailed picture of the phenomenon. But things are different now.

The problem was winter. The lack of sunlight illuminating Saturn's north pole would keep it in the dark until 2009, when the gas giant's spring season came around and allowed us to get our first detailed look at the enormity of the hexagon.

While the interior of the hexagon remained a mystery,the air current that makes up the hexagon is quite clear, as well as the sheer number of hurricanes and storm systems that surround it in perpetuity, like this one that Cassini filmed a year ago:

However, in late 2012, light began to illuminate the pole's interior, and the probe has been able to observe the hexagon even more clearly than before. Taken over a 10-hour time span with high resolution cameras, NASA was able to obtain a clear, moving image of the planet's distinct air current and its surrounding storm systems. The images were then analyzed through a rendering method knowns as false color, which allows the different types of particles suspended in the atmosphere to be easily identified.

The view will only get better as we approach 2017, Saturn's summer solstice--and Cassini's projected end of life. It's a great end to a year full of breathtaking images taken by the Cassini, like this past summer's "Day the Earth Smiled," the first time NASA informed the world that it would be taking a picture of Earth from space in advance. Here are some great ones:

In order to capture these images, the Cassini makes use of its Imaging Science Subsystem (ISS), a remote sensing instrument that can capture images in most visible light, along with some infrared and ultraviolet capabilities. The instrument is outfitted with both wide- and narrow-angle cameras that are equipped with charge-coupled devices for use as an electromagnetic wave detector. Per Wikipedia:

"Each CCD has a 1,024 square array of pixels, 12 μm on a side. Both cameras allow for many data collection modes, including on-chip data compression. Both cameras are fitted with spectral filters that rotate on a wheel--to view different bands within the electromagnetic spectrum ranging from 0.2 to 1.1 μm."

How The Fourth Dimension Of Sound Is Being Used For Live Concerts

$
0
0

Max Cooper wants you to experience music in four dimensions. The electronic musician has been experimenting with a new sound system that uses several omnidirectional speakers and special software to deliver audio in a more immersive, interactive way. We talked to Cooper about 4-D sound, its implications, and how he sees it being used in the future.

What's the reaction you get from people when explaining 4-D sound?

People are very interested when I tell them about it. They don't need much convincing once I've explained the basics. And just to clarify the name--the system is called "4DSOUND" because the sounds exist not only in three dimensional space, but they also move in time, the fourth dimension.

Is 4DSOUND all about the experience, as opposed to the music?

It certainly is a new way of experiencing music, but I don't think that experience alone would carry a show. No doubt some people might come just to hear the new format, but if they didn't like my music in the first place then they probably wouldn't like the show as a whole--it's an addition to the music rather than a replacement.

If you look up "music" on Wikipedia, it is defined by pitch, rhythm, dynamics, timbre, and texture. The 4-D system literally adds an extra category of three dimensional spatiality to the list of what music is. Each piece of music turns into a physical entity that can be explored and interacted with, as it envelops and moves around you.

Do traditional live shows and audio recordings feel stale to you after experiencing 4DSOUND?

I am really excited about the 4-D project. It opens so many doors musically that I'm only just starting to explore. But it doesn't replace or nullify traditional shows--both have pros and cons. The standard 4-D system is built of 48 omni-directional speakers and nine subs. The omni-directional speakers project sound in all directions, allowing sounds to seem to come from anywhere in the space, or even outside of it. Where normal club systems are directional and punch you in the chest from a specific unchanging position.

What's the learning curve for creating music in this spatial way?

The developers of the 4-D system have created a great interface for Ableton's Max for Live system, which provides a big range of spatial controls specific for the 4-D rig. But it does take a while to get the hang of it, a lot more preparation needed than for a standard show, but there's a lot more potential too.

How likely are venues to actually implement this speaker setup? What can be done to get people on board faster?

At the moment it's very much an experimental system, needing a lot of expertise and work to make it happen. It also needs a very big space--at least 11 meters square by 6 meters high, and ideally with a much larger floor space than that. It's something for galleries and installation spaces as much as clubs, and there's been a lot of interest from festivals.

I don't think anything should be done to get people on board faster. People are really interested already from what I've seen, and it's a matter of continuing to develop the technology and its artistic use organically so that the more it spreads the higher quality the results are. It's still early stages. There is a lot of work to be done before it's ready to take over the world. Also, it needs two trucks to transport all the kit at the moment, which makes it expensive to stage a 4-D show.

Is the 4-D sound system about trying to increase interest in the live show over home consumption?

Nothing compares to a great club experience in a lot of ways, unless you want to totally destroy your living space, so it's not about competing with home experiences for me. I would agree that for some forms of media, like movies, it's now hard to beat the potential home experience. But can clubs go down that route? I don't think so.

The 4-D show is a club experience, but it's also almost a museum experience. It's somewhere in between, which has always been an aim for me musically. Some of my music is called "techno" for example, but it's just as much for home listening even though it's of a club format.

I want to make music that isn't just for hammered people in clubs. I want it to have some real musical quality that can be enjoyed at home in a sober state as well.

What do you see for the future of live shows?

Home viewing/listening is great for enjoying music and performances on a certain level. But the live experience provides so much more than just the information content of music piped into your home. The information content of the environment is possibly more important to me. My favorite clubs and festivals, for example, aren't those where my favorite artists play. They're places where I'm enveloped by an amazing new interactive experience that's unpredictable on every level.

Home entertainment is probably 100 years off being able to deliver something like that in my opinion--it would have to be a direct link to the brain. Go to Glastonbury festival in the U.K., or Wilde Renate or Berghain in Berlin and compare that to any possible home experience. And people can't get that at home, so they will always want to go out!

Since part of the benefit of the 4-D system is being able to record sounds as they naturally occur in space and time, what are some of the practical use cases you've encountered?

Yes, our delivery of sound and other media always has an ongoing mission to make things more natural, because the natural world is always the richest source of experience. Nothing compares to the real thing. The 4-D system allows the creation of very unreal, unnatural sound environments, which is fun--it provides a surreal, but real sound environment.

One practical issue we've been considering with 4-D is whether we can properly translate the experience to home listening. I recorded my live show with binaural microphones to try and achieve this. They are mics which record from inside the ears, so that they capture the effects of the shape of the ears and transmission of sound through the bones around the ear, within the recording.

This means, that, in theory, when someone listens back to the recording with headphones, their brain picks up the same signals which tell it about the surrounding environment. It can be pretty convincing, and binaural recording is surprisingly popular. Demos like "the virtual hairdressers" on YouTube have had as many as 16 million hits.

We've recorded the 4-D show using that same technique, and it sounds really interesting. But it's not as powerful as feeling the sound live, and you can't walk around and explore the environment in the same way. Maybe we could create an interactive app, but that's a ways off!

The Wisdom Of The 20-Minute Startup

$
0
0

"What cool new products are you using?"

We all ask this question. It's a common conversation starter, especially in the startup community. I'm particularly fond of this topic--I enjoy geeking out about products, writing design deconstructions, and swapping discoveries with smart folks. But these conversations provide more than just entertainment value: They are also a great learning opportunity. Understanding the subtleties of good and bad products is critical for product builders. As Paul Buchheit says, you must "live in the future" to shape it. Playing with early, innovative products can provide a competitive advantage.

This was the basis for Product Hunt. Here's how we prototyped it.

The Idea In Its Simplest Form

The concept was simple: to build a community for product people to share, discover, and discuss new and interesting products. But when I came up with the idea, I lamented the amount of work needed to build a first version of Product Hunt. Even a basic Ruby on Rails app would take me weeks to build. Although confident in my idea, I didn't really know if anyone else would use a service like this. I wasn't about to spend dozens of my nights and weekends building something no one cared about. How could I bring it to market sooner to test my hypothesis?

The 20-Minute MVP

It was unusually chilly that morning in San Francisco when I walked to my office, Philz Coffee. I ordered and claimed my usual seat. After unloading my MacBook, I peeked at my to-do list to find something I jotted the previous week:

  • Create Product Hunt

In a burst of motivation to make Product Hunt a reality, I brainstormed ways to build a quick MVP to see if people cared to share and discover products. After noodling over a few ideas, I was reminded of Linkydink, a link-sharing tool by the friendly folks at Makeshift. Simply create a group and invite people to share links with other contributors and subscribers. Each day, the collection of posts are emailed to the group. "This is perfect!" I thought, mentally fist-pumping with excitement.

I logged into Linkydink, created a group, and invited a few of my startup friends to contribute. I wrote a quick blog post, announced the project on Quibb and tweeted.

Within 20 minutes, I had an MVP.

I sat back, sipping my coffee, anxious to see how people would respond.

The Results

Immediately, I received overwhelmingly positive feedback, first from Ash Bhoopathy, entrepreneur-in-action at Sequoia Capital.

Fun experiment, I've been wondering why something like this hasn't existed on a larger scale. It's probably the PM / growth dude's best replacement for 'Dribbble' for collecting inspiration.

Then from a partner at Union Square Ventures, Andrew Weissman:

Love how you have all kinds of VCs subscribed! Build an angel list syndicate off this list and disrupt them (us) ;-).

And then from Talton Figgins, product support lead, Disqus:

Wrote this idea off at first when I first read about it but after checking out some of the recommendations (Peak, Sqwiggle, Calm, and Cycloramic) I'm hooked. Can't wait to check out more.

Within two weeks, over 170 people had subscribed to product discoveries from 30 hand-picked contributors, consisting of startup founders, VCs, and prominent bloggers. Even more encouraging were the numerous unsolicited emails and in-person conversations expressing their love and support of the project.

It's still very early but these signs of traction are encouraging, especially considering the minimalism of the "Linkydink MVP" and my (intentional) lack of marketing of the product. Granted, I didn't launch the MVP with a blank slate. Years of blogging, relationship building, and projects like Startup Edition have given me an audience and network of supporters. The term "startup" is deceiving. Successful companies don't start up overnight; they are founded upon years of experience and help from others that must be earned.

Now, For Building a "Real" Product Hunt

The results of the MVP gave me confidence in the idea: I had found something compelling. I began to research technology to build a complete product. Some of my engineering friends recommended Sinatra or Ruby on Rails. Unfortunately, I didn't have any experience with these frameworks and while assured in my idea, I was still operating with many hypotheses of what the product could be. I wanted something sooner to test my next series of assumptions. I started to look into Sacha Greif's Telescope, an open-source app for creating your own Hacker News or Reddit-like community. Pleased, I reached out to my good friend and designer/developer Nathan Bashaw to get his thoughts:

Yes! Eight days later, we launched Product Hunt.

An MVP is Not a Product

The "20-minute Linkydink MVP" was a great starting point. It allowed me to validate some assumptions very quickly and observe real user behavior without a single line of code. Entrepreneurs often assume an MVP needs to be "built." The purpose of an MVP is to learn, to validate and invalidate assumptions. There are almost always faster ways to do this than building a product.

Next time you have an idea for a new product or startup, ask these questions before touching a line of code:

  1. Identify your assumptions: What assumptions do you have about your idea? What must be true for the product to succeed? How do you identify your audience?
  2. Test your assumptions: With your riskiest assumption in hand, brainstorm ways to validate or invalidate that assumption quickly. Use landing page tests, Wizard of Oz experiments, customer interviews, and other tactics to gather feedback from your target audience.
  3. Learn and iterate: If your assumption in the previous step is invalid, reevaluate the idea and your target audience. If you're lucky enough to be right, congratulate yourself and test the next riskiest assumption.

This lean approach and mentality was used to birth Product Hunt and its ongoing development.

Join the Hunt

Over Thanksgiving break, my talented friend Nathan built the first version of Product Hunt. On November 26th we began seeding the community with a few select founding contributors. A week later, we formally launched in the press, receiving another healthy spike of adoption and feedback.

This is just the first post of a series where I will share the strategies, tactics, and surprises we encounter building Product Hunt. Check back on FastCo.Labs for more, or subscribe to my blog to follow along.

Ryan Hoover is the director of product at PlayHaven and creator of Startup Edition. Visit his blog and follow him on Twitter for more on startups and product design.

Bringing The Maker Revolution To Africa, One Raspberry Pi At A Time

$
0
0

Is tech education what the developing world needs? As the argument goes, what good is learning to code when water purification is still a fantasy? One answer has come from HacKIDemia's Afrimakers project, a crowdsourced campaign to bring the "maker revolution" to the third world via a kit of sensors, microcontrollers, and good ol' fashioned scrappy programming.

Afrimakers wants its communities to take the program-donated box of goodies to "use and create extremely affordable technology for solving local challenges." To that end, their goody box includes a range of sensors for detecting things like humidity, temperature, pulse, color, and carbon monoxide to hook up to Raspberry Pi and quartet of Arduinos in each kit. With a few breadboards for good measure, these sensors can monitor local conditions or be put on a wearable rig for LED-linked health monitors.

But Afrimakers doesn't just want to airdrop hardware components. Integral to the project's mission is the hands-on training to build "hubs" in seven African countries--Kenya, Tanzania, South Africa, Zambia, Ghana, Nigeria, Egypt--intended to become regional tech education centers of their own after HacKIDemia trains local volunteers to become resource experts.

The kit-equipped hubs will host classroom workshops for kids to immerse themselves in the immediate-reward tangibility of maker tinkering--skills that will transfer to constructing gadgets and resource monitors that the Afrimakers founders hope will improve local quality of life.

The above video is from the Afrimakers pilot program in Nigeria. These 400 kids did much more than light up an LED with a potato. Their list of accomplishes includes building a waste-to-energy converter, DNA extractor, and robotic mini-backhoe constructor.

The Afrimakers Indiegogo campaign has raised $8,432 of its $50,000 goal--enough to fund two of the planned seven hubs, each of which will get two kits costing $1,000 apiece. Each hub's team of volunteer experts has a self-financing campaign charging wealthier private schools for workshop classes, which in turn funds the free workshops that the teams bring to public schools.

It should be noted that HacKIDemia, the youth-maker-educating group behind Afrimakers, has been arranging accessible workshops for youth around the world since Summer 2012. With a board of mentors like astronaut Dan Berry, founder and director of Fundación Escuela Nueva Vicky Colbert, and Stanford lecturer of Entrepreneurship and Clean Energy Tony Seba, HacKIDemia has established 23 hubs on five continents.

This Guy Made His Own iBeacon For $70

$
0
0

Last week, Apple became the latest retailer to use its own iBeacon technology in a bricks-and-mortar retail environment. But the tech--which enables stores to beam messages to iOS device-toting patrons--is based on the existing low energy Bluetooth spec. So is there any reason you couldn't just build your own with a Raspberry Pi and Bluetooth dongle?

Nope! You can build your own indeed, and that's exactly what Tony Smith did recently. The El Reg writer made the equivalent of an iBeacon with a $50 Raspberry Pi and $20 Bluetooth 4.0 dongle, prepped the Pi with the Linux Bluetooth stack, BlueZ, and various USB development packages. Any curious tinkerer can follow Smith's instructions to set up the beacon, then head to this Radius Networks write-up to script one's own advertising message. Tada!

The iBeacon beams up to 31-byte packets to "participating" iOS devices--in Macy's case, anyone who downloaded their Shopkick-designed app. Each packet includes a 128-bit Universally Unique Identifier (UUID) keyed to the service (IDing the packet as Macy's, say) along with two 16-bit numbers, a "Major" number (which might distinguish stores) and a "Minor" number (which might distinguish departments).

At least Macy's is using a voluntary app to track you and provide useful information as it tracks your metadata: Previous tactics, like tracking your phone via its autobroadcasting Wi-Fi signature, are more nefarious. As we've previously pointed out, iBeacon's microlocation services also allow for local authentication--which, combined with the iPhone 5S's fingerprint scanner, has allowed Apple Store shoppers to scan and buy products through their iTunes account without ever talking to a store rep.

Having your own little iBeacon setup means that these shop features could be yours without running through Apple's stringently closed product channels. Whether that amounts to pinging your iPhone-equipped kids with an annoying pop-up when dinner's ready is up to you.

The Revolutionary New Music Apps You Missed In 2013

$
0
0

A lot of new music services continue to pop up each year, despite most consumer's unwillingness to financially support artists directly. This is why most music services practice the balancing act of end user visibility and technical improvements to existing ideas, in order to generate both buzz and money. The music apps and services listed below won't necessarily revolutionize the industry, but they are some of the most interesting to debut in 2013 and worth a first, or second, look.

Bop.fm

As different streaming services pile up, Bop.fm is trying to connect the dots between them. Instead of alienating your Rdio friends with Spotify links and vice versa, Bop.fm will generate a universal link that connects paid streaming services as well as SoundCloud and YouTube for your freeloader friends.

Sharing music has always had a high barrier to entry, including the analog method of writing recommendations down on paper. Content resolution is probably a back-end issue long-term, but in the interim, it's exciting to see some more easy-to-use movement in the area.

Announced today, Bop.fm will be partnering with the streaming service Deezer. Until the music service eventually comes to the U.S., Bop.fm will allow its users to share tracks from the service that resolve to different music outlets for U.S. users.

Also see Tomahawk--an open source alternative that's been around for a while.

Mindie

At first glance, Mindie looks like a Vine competitor, but with music tacked on. The humble app is a way to create an instant soundtrack and set the mood for the video's viewer. But from a big picture perspective it's also conceivable that Mindie is a chance for artists and fans to redefine the modern-day music video. Instead of needing to share the whole song, sharing the song's hook along with some visuals may be all an attention-starved generation needs to discover new artists.

Mindie is in the process of making it over the user adoption hump as public figures like Aston Kutcher, Michael Arrington, and even Twitter's Jack Dorsey have all been sharing clips through the service.

Upbeat

Ever since Digg hit on a winning formula for ranking Internet content, the technique has been adopted into every category and use case--music included. While there have been others to do the same or similar things, Upbeat finds a pleasant way to rank user-submitted music. Laid out in three simple columns, Upbeat is exactly what you might hope for in a music chart for the Internet age.

You can queue tracks, save them, and purchase them on different services. If you've been in a new-music funk and need something to shake it, chances are you'll find some new songs on Upbeat.

DistroKid

Until Apple offers musicians the same direct access to iTunes that developers have for the App Store, third-party solutions will be critical for independent artists. DistroKid is a simple service that places an artist's music in different online stores for sale, including Amazon, Google, and Spotify. As simple and easy to use as the service is, the pricing is equally as easy to understand.

The service offers the ability to upload a song for free, no credit card required, as well as get unlimited music into the stores for $20/year. Seems obvious enough, but DistroKid's pricing is part of what makes it so interesting. Artists retain all their rights, don't pay a percentage, and don't have to worry about other types of scamming practices. The service can operate affordably thanks to a lot of back-end automation.

Jukely

Finding concerts in your area has become a piece of cake with plenty of apps that scan your music library and know your tastes instantly. Jukely approaches the concert from a different angle than most before it, however, and takes on a promoter's interests. Jukely wants to find the best concerts for you, at the same time hand-picking venues that host the best shows.

As concert venues head online, there's more of a need to get people to come to live shows and support touring artists. Jukely incorporates hand-picked curation in addition to back-end algorithms to deliver great picks right to your phone.

Turntable Live

Turntable.fm--the online music listening room--finally crashed and burned, but out of the ashes came Turntable Live. Aiming to be the future of concerts, the site broadcasts live shows from its NYC office, charging viewers a small fee which most of goes to the band performing.

It depends on who you ask whether this model will work, possibly angering those die-hard concertgoers, but it's the best effort to date. There are a lot of threats and potential for online concerts that charge money, but very few beyond Turntable Live has actually made an effort.

Whyd

Whyd is like Facebook, if Facebook was exclusively for music and wasn't the butt of every joke. On Whyd you get a stream of new music being added and collected by people you follow, being able to see things like which playlists they add them to. For music enthusiasts head-deep into searching for new music, one of the most helpful things is a community of people doing the same thing, and Whyd facilitates that.

What makes Whyd interesting beyond a solid experience for music nerds is the bookmarklet which scrapes songs from most sites and allows you to track, save, and remember artists you find intriguing. The tool isn't perfect, but it works as well as you'd hope.

While the service is still in beta, those interested can use this link to gain entrance to the service.

iTunes Radio

When iTunes Radio launched with iOS 7, it meant that every iPhone, iPod, and iPad now sold comes with a built-in radio function. Beyond the tired debate comparing Pandora or other radio services, just having a new device from one of the world's biggest manufactures come loaded with direct access to unlimited music, for free out of the box, is a big deal.

Twitter Music

Unlike the other apps and services listed, Twitter #music is interesting for a host of depressing reasons. No one could have guessed that in the same year the app launched that it would basically be scrapped for parts. We haven't seen the last of a music push by Twitter, but when it does refocus, it'll look a lot different than it first did in March of 2013.

If You Want To Hire Developers, You Better Hire Fast

$
0
0

Startups are constantly competing with bigger, wealthier companies for talent. And they often win out, even though seed companies pay less on average than corporate development gigs.

Matt Mickiewicz runs developer recruitment site Hired. He says the advantage for startups is their speed. Hired's in-house data scientist Elliot Kulakow analyzed nine months of data across 2,300 hiring processes and shared the data with us to show us how the speed factor works in tech recruitment.

Top Talent is Hired Faster

Eighty-three percent of paper offers are accepted if they are made within 21 days of an introduction, the point at which a company and developer agree to start a formal interview process. After that the percentage drops to 63%. Speed is even more critical when it comes to hiring the most in-demand developers. Two-thirds of offers go to the top third of talent and developers who receive eight or more offers on Hired accept a job within 16 days while those who receive a single offer take over 23 days.

How Fast Do Companies Hire?

The fastest companies average four days between each interview step, the slowest 10 days. That means that the fastest quartile of companies complete a full four-step interview process before the slowest quartile have even gotten to the third step.

The more steps there are in your hiring process, the more likely it is that you will be too slow to hire top candidates. "You should be able to present a final paper offer within 5-10 days of first meeting someone," says Mickiewicz. "Move the process as fast as the candidate allows versus as fast as is convenient for you. We have seen hundreds of cases where companies just forget to follow up with candidates or to reject a candidate. They'll reschedule interviews three times because of other priorities."

Startups Versus Corporations

Seed stage companies can often make faster decisions than larger rivals, and this shows in the statistics, but the first quartile of series A and B+ companies and even major corporations are still faster to make an offer than most seed-funded startups. Series B+ companies hire more people--they are almost three times more likely to make a final offer than seed stage companies--so their hiring process tend to be much more organized than those of early-stage startups.

Later-stage companies also make better salary offers. Seed stage companies pay, on average, below market rates but the difference is most pronounced for senior developers. While a senior engineer can expect around $135,000 at a larger company, seed companies typically offer around $115,000.

Early-stage companies may also lose out by underestimating how much time they should spend on recruitment. "Twenty to twenty-five percent of your time should be spent interviewing," says Mickiewicz. "It's a really good metric as to whether you have a hiring culture. If you view hiring as a core competency you need to develop in the business, then you'll do whatever it takes."


Could The Road To The "Dark Web" Be The Right One?

$
0
0

The NSA's secret surveillance program raised a number of issues regarding one's ability--and right--to remain anonymous on the Internet. For the web-savvy, navigating outside the purview of a corporate- and government-monitored Net required a special kind of web browser called Tor.

Tor has gotten something of a bad rap as a tool for guns, drugs, or human trafficking, and indeed many of the people that use the Internet anonymously may have things to hide. But Andrew Lewman, executive director of the Tor Project, says now's the time for everyday users to lose their trepidation and start browsing anonymously, no matter the present company.

An Internet For Drugs, Child Pornography, And Hitmen?

"No. I'm not into kiddie porn."

That's what a friend of mine says when I ask him if he's ever used Tor.

When I ask another friend, she says, "That's 'The Dark Web,' right? That's Silk Road where you can buy guns and drugs and stuff? No, I've never used it, but I don't need to have anyone killed right now."

And, like my friends, if you have heard of Tor, which is also known by the shady monikers of "The Hidden Web," "The Deep Net," and "DarkNet," you could be forgiven for thinking "Tor" equals "hangout for murderers, child pornographers, and drug dealers." You could also be forgiven for linking Tor with guys like Dread Pirate Roberts, the operator of Silk Road, the online marketplace for guns and drugs and trafficking victims.

But that wouldn't quite be fair, says Lewman when I tell him my friends' impressions.

"Does your friend use a smartphone? How about cars? Kitchen knives?" Lewman says. "Then he's using the same tools as child abusers, terrorists, and other criminals. Technology is agnostic--the morality is in the human using it."

The misconceptions about what Tor is are enormous. Tor is not a website, nor is it affiliated with the Silk Road, which was shut down by the FBI this October. It's more like... the safest road to the Silk Road.

"Silk Road is a funny situation," Lewman says. "In some circles, it's a horror show of how drugs have permeated our culture. In other circles, it lets people safely get their drugs without having to resort to sketchy people in sketchy parts of the city."

Many people are surprised to hear that the Tor Project is an authentic nonprofit that gets its funding from, among other sources, the U.S. and Swedish governments, not to mention that it originally grew out of an onion routing project run by the U.S. Navy.

Tor, like life--and the Internet we are all used to--is what you make of it. On the client end Tor is a simple browser that runs on most desktop operating systems that allows users to access a second Internet. This is a peer-to-peer Internet of anonymous, interconnected computers that make finding the true location or identity of a user virtually impossible. But while it's true that this anonymity has led some people and organizations to set up shady websites that allow you to buy all kinds of illegal wares, that's only because those people chose to use Tor to do that. Another group of people could very well choose to make a Tor website that delivers free puppies to lonely orphans.

"The hidden Net is no different than the Internet in the early 1990s," Lewman says. "The original pioneers using the Internet were criminals and pornographers trying to figure out how to make money with it. And then came everyone else. This is the same situation we're in now, where early adopters are exploring the hidden networks and figuring out how to make money or just exist within it."

But those early adopters many not be alone for much longer. While for many a second Internet may have seemed like an unnecessary complication just a year ago, the revelations of the existence of the NSA's PRISM program and the reach governments now have into hacking virtually anyone's personal information have brought the idea of a new Internet that, first and foremost, protects an individual's privacy and anonymity to the forefront of the "average" user's mind. PRISM also serves as a cold wakeup call for anyone who's ever thought privacy has existed on the web we've been using since the mid-'90s.

On The Internet, Everyone Knows You're A Dog

In 1993, the New Yorker published a cartoon by Peter Steiner that showed a dog sitting at a desk in front of a computer. The dog at the desk is addressing another dog sitting on the floor. He says, "On the Internet, nobody knows you're a dog."

The cartoon was the New Yorker's most reprinted cartoon ever and conveyed the sense of total anonymity users felt they had on the then-new global communication platform that would carry the world into the 21st century.

The only problem, according to Lewman, was the New Yorker's cartoon was a fallacy.

"The Internet is not private. It never has been. Just because you don't use your name doesn't mean you aren't giving up intimate details of your life just by browsing around. The New Yorker furthered this misconception," Lewman says. "In fact, everyone knows you're a dog, what you like to read, buy, and post, and where you live. They then correlate that with offline data and have a complete picture of you. The 'they' can be advertising, marketing, surveillance, security, or e-commerce companies. Then there's the government-related 'they' in law enforcement, intelligence agencies, et cetera."

I think it would be foolish to assume that in 2013 most people don't think corporations like Facebook and Google and Amazon aren't tracking our anonymized Internet presence--where we go, what we click on. Taken to the next level, even when it became public knowledge that the U.S. government had the technology to spy on all of our online activity at any time, the public outrage quickly died down and people returned to their busy lives. The average person's attitude became, "So what? What am I going to do about it?" or "Why should I care? I'm not a terrorist."

But people should care, because the lack of Internet privacy and security cannot only be used against terrorists and those doing harm in the world. It can be used against political dissidents, journalists, and activists--in other words: people who generally work anonymously, for a period of time, to make the world a better place.

And that's why Tor, the perception people have of it, and the continued expansion of its user base is so important. It allows political dissidents, journalists, and activists--many of whom use Tor just as frequently as arms and drug dealers do--to communicate safely, securely, and anonymously.

But the fact is most of us aren't political dissidents, journalists, and activists. Perhaps that's why the public outrage over PRISM has died down so much, because many are at ease that a "good" government (and I do realize that's a very relative description) like the U.S. is in charge of PRISM. But if the U.S. can do it, China can do it. And in the future many other less savory governments will be able to look into their citizens' lives at any time. And that's a very bad thing. After all, who's to say there won't be another controlling party like the Nazis in a hundred year's time running the U.S. or any other country? How many more Jews would have died in Nazi Germany if the Internet was the communication tool of choice back then and the government could see into anyone's lives at will? What if a government of the future decides anyone who has ever bought a left-leaning book from Amazon is a political threat and needs to be rounded up and eliminated?

The Future Of Internet Privacy

What a future evil government could do with the technological capabilities of PRISM is, of course, theoretical and perhaps best left to dystopian thinkers. However, just because some implications seem farfetched doesn't mean technology like Tor that could make us less surveillance-prone shouldn't be explored by the general public.

But for all the anonymity protections Tor has built in, that doesn't mean it's the final solution to anti-PRISM security, something Lewman readily admits.

For true anonymity, Lewman says users shouldn't solely rely on Tor; they must also be willing to change a number of their Internet usage habits. He also says that governments aren't the only threat, noting that no matter what the public face of a company proclaims, the big tech giants should not be trusted. "Unless it's free software you can have examined, one shouldn't trust them. People need to understand what they're giving to third parties when using them. Third parties could start being far more transparent about what they do with your data. Google is leading the way here, but everyone still has a long way to go."

As for mutual trust, I ask Lewman if any government agency has ever asked him for a back door into Tor. He tells me "no" and says that if any government did ask, his lawyers are confident they could fight the request in court and win. As for the NSA attacking Tor directly, Lewman says that despite a leaked NSA slide from 2007 saying the agency had no way to de-anonymize Tor, he can't be sure that's still the case. "No idea," he says. "Likely, no. We know they are trying to attack the browser and the user, not the encryption and Tor itself."

These constant back-and-forth battles over government surveillance, along with constant pressure from technology companies, many of whom's business plans rely on us sharing our likes, purchase habits, and even feelings, makes the future of anonymity on the Internet seem bleak. But Lewman says it's important people don't give up.

"Anonymity has always mattered because anonymous speech is critical to a functional democracy," he says. "As speech moves online, so should the anonymity, if needed. People who want to sharecrop for large corporations can make their own decisions. The ability to have a choice to speak in your name or not means we're still doing something right."

As for the future of Tor, in light of the NSA revelations and the social sharing overload many of us are feeling, does Lewman feel that the world is ready for a second Internet, and do we (adapting Thomas Paine's words for the digital age) have it in our power to begin the web over again?

"Yes," he says. "Between the copyright cartels, data aggregation companies, and now governments spying and stalking on users, people are already looking for alternatives. We're at the beginning of a new wave of privacy-enhancing technology coming into the mainstream. The first ones to make it easy, sexy, and profitable will lead the way."

Why Upworthy's Signature Headlines Are Only Half The Story

$
0
0

A year and a half after its launch, it seems as though Upworthy is legitimately onto something. On a superficial level, the reason seems obvious: The company's distinct headline style, which is both widely parodied and imitated, is working like gangbusters--the company is rapidly closing in on traffic giants like BuzzFeed and Gawker. But below the surface is a story of social networks and the algorithms they tinker with, and how tweaks to the formula can upend even the best laid plans for reaching massive amounts of people online.

In a piece for the Atlantic, writer Robinson Meyer argues that while Upworthy and its many imitators have found success with their signature style, it was just a part of a larger story: Upworthy is one of the smaller traffic kingdoms in the middle of a distribution battle between Facebook and Twitter.

Meyer cites this Facebook corporate blog post from mid-October in which the social media giant states that average referral traffic it has shuttled to other media sites has increased by 170% in the past year. Last week, Facebook explained why: it's actively trying to promote what it calls "high-quality content" in user's News Feeds, and will now begin to bump them up even more than before.

While no one really knows precisely what makes Facebook's News Feed algorithms tick, Upworthy's approach is to specifically design its pages, buttons, and layouts to be Facebook shareable. The company claims that their headlines aren't what matters most, it's that people share their posts.

Whatever tweaks Facebook made to News Feed helped make Upworthy the new king of the viral content game: From October to November, their unique monthly visitors absolutely skyrocketed from 42 million to 88 million.

As Facebook and Twitter reach something close to critical mass, new users become much less of a reliable metric for their success. Instead, they both want to be the feed you scroll through most. Sharing and re-sharing is now the means by which any content provider worth its salt will define their worth.

What separates the big distribution networks (Facebook, Twitter, LinkedIn and Tumblr) from an intermediary content-provider like Upworthy is community. There's no commenting on Upworthy--it doesn't want a community of its own.

It has been argued by some that the tremendous growth advantage provided by the perfect storm of Upworthy's viral strategy and social networks' viral ambitions may not be an ethical or healthy practice. As part of a long essay over at Gawker, Tom Scocca argues that it is symptomatic of a larger problem of the way we communicate online. Namely, that it's limiting: To go viral the Upworthy way, to be spread and shared and tweeted about, you have to be liked. You have to be nice. You can't be divisive or mean or disagree. Who wants to invite a buzzkill?

"The result of this approach, the Upworthy house style, is a coy sort of emulation of English, stripped of actual semantic content: This Man Removed the Specific and the Negative, and What Happened Next Will Astonish You," says Scocca.

How Much Of Facebook's Footprint Is Thanks To Parse?

$
0
0

Today at the LeWeb developer conference in Paris, Facebook released a heat map of developers working on Facebook apps all over Earth. While the map shows the social network's significant adoption among developers in much of the world, it only hints at the ways the company is planning to expand its development community.

Click to expand

The image marks the first time Facebook has released any numbers on its developers since its acquisition of Parse--a company that provides cloud-based backend services (or BAAS) for app developers--way back in April.

Parse has also been relatively quiet about the size of the development community using its service. When it officially launched in the spring of 2012, the service had 10,000 developers on board. In one year's time, it was bought by Facebook--which caused even more developers to flock to the service. How many, they wouldn't say, but a June update revealed that 100,000 apps had been made using Parse.

Speaking to VentureBeat last September, Parse CEO Ilya Sukhar wouldn't say much more about the company's figures on developers using the service, but he did give the clear impression of rapid growth following the acquisition:

"Our daily signups are many multiples of what they are before we were acquired." Shortly after the acquisition, daily signups had spiked to 10 times the pre-acquisition average, causing the company to gain 20,000 new apps in just over a month. "We have much more visibility and much more momentum" with the Facebook brand attached, he said. "We're really happy with it."

Parse's rapid growth following its acquisition would suggest that, if Parse isn't at least partially responsible for the numbers on Facebook's newly released chart, it's a big part of the network's push toward attracting more developers worldwide. Mike Randall is the global head of Facebook's Preferred Marketing Developer (PMD) program, an internal group that recruits developers to work with select advertisers in order to optimize engagement across Facebook's entire platform. In an interview with The Next Web, Randall says that his program is one of the key areas in which Facebook aims to grow it's developer pool:

"I like to bucket them in three areas of where we're being more prescriptive, for lack of a better term. One is on the business front, continuing to push for global expansion--in some cases that might mean a PMD opens an office in a particular country [or] it might make more sense for that PMD to partner with a local company. We'd love to see a combination of those companies that operate at scale, but also local stars that rise up from the culture, whether that's in Brazil or Singapore or Australia."

An asset like Parse, which would have all manner of analytics on the developers that use it for Facebook's perusal, would be instrumental in identifying developers suitable for its ambitions at the intersection of development and marketing. It could also give the social network the means by which it could identify the next big startup for it to acquire.

Feedly Hijacks Your Links, But This Time It's Unintentional

$
0
0

Since the death of Google Reader back in July, RSS feed aggregators like Feedly and NewsBlur have filled the void, working to replicate the features--and ethics--of Google's service. Yesterday, somebody discovered that Feedly was linking in-feed sponsored content links back to its own site, sparking a backlash on Twitter. But was it intentional?

The story begins, as so many do, with a tweet to former lead Tumblr developer and Instapaper creator Marco Arment informing him that a sponsored link from his personal blog, when viewed on a Feedly reader, sourced the sponsor-funded click back to Feedly instead. Arment tweeted his dismay and a storm of anti-Feedly rage swarmed over Twitter. A little over an hour later, Feedly tweeted out a fix with a blog post explaining what was really happening.

Very basically, Feedly had responded to publisher requests this summer (when Google Reader died and Feedly absorbed publishers in need of professional tools) to track traffic generated by Feedly readers by adding a "utm_source" decoration to the title of articles which leads users from Feedly back to the publisher's site.

But this messes with article titles linking back to sponsors who track referrals from blogs like Arment's using the "utm_source" decoration. In Arment's case, "https://app.io/?utm_source=Marco.org&utm_medium=blog&utm_campaign=blogpost" was autochanged by Feedly to "https://app.io/?utm_source=feedly". Instead, Feedly will slap a different decoration on the end of titles, "utm_reader=feedly", and have publishers switch over to tracking the utm_reader decoration. Which is great, but this change was made in the summer--meaning an untold amount of revenue from this kind of tracking is irretrievably lost.

This is the latest decision by Feedly to leave publishers pissed. Last Friday, an unannounced update switched articles shared from within a Feedly reader to redirect to a copy of the article hosted on a Feedly server. While that post updated within the day to note that Feedly had reversed its decision, it raised, for the millionth time, the specter of who-owns-what in the RSS game.

And it's not going to stop. David Smith, creator of RSS reader Feed Wrangler, posted a manifesto of what ethics RSS readers should apply by--and he posted it yesterday, half a day before today's Feedly brouhaha. It would be technically easy to scrape content, Smith says--it's his choice to respect truncation and formatting choices of the publishers that counts."The value of a service like Feed Wrangler is entirely driven by the abundance of quality content that my customers can subscribe to," Smith said. "If I'm making choices that make it harder for writers to make a living I'm ultimately just shooting myself in the foot."

The Ultimate Excuse Generator For New Yorkers

$
0
0

As a few of you know, the MTA has a "delay verification" to print off subway delay confirmations. But let's say you want to abuse the delay verification service and find the slowest routes to your destination--presumably as an excuse to miss something you don't want to go to. Some company Christmas party you're dreading? Some messy family gathering you'd prefer to nap through?

Well, the official excuse verification system is inadequate because you have to do it after the fact. The web tool, the excuse generator, was introduced in 2010, but verifications can take hours or days to return. The New York Times followed up on the service's progress, noting its nominal utility. If you're living in a world where lateness matters, like school, or court, or work, it's probably punishable on the spot. This is what you'd call a sub-optimal user experience, since you have to generate the excuse after the fact. If you want a truly useful excuse generator, you'll have to drink directly from the firehose.

Behold the MTA's real-time feed for delays and changes, which is free to use and updates every minute. Hitting this feed results prints all the delays in the city subway system with descriptions, as they evolve from minute to minute. In its municipal mercy, the data comes in .xml, which means any XML reader plugin will make sense of the madness--just search for the path "/service/subway/line/name" and find the line of your chosen subway. XML illiterate? Just CTRL+F for your line group (such as "ACE" or "NQR"). The status field will even tell you if you're facing delays or planned work. Then your alibi is solid.

At the moment, the delay verification only works for trains, but who knows when the verification will be extended to the colossal 230-line, 5,700-bus NYC bus system.

How PayPal Created A Private Cloud And Switched To Agile

$
0
0

PayPal is one of the biggest financial institutions in the world, with 130 million accounts and more than 13,000 employees, but in recent years the company's success has become an obstacle. While a new crop of payments startups were nipping at PayPal's heels, the company was moving more slowly than ever. Development teams were still following a waterfall model, not the more modern agile development, and launching even a small service required more than 100 tickets and it took three weeks to provision 50 servers. To fix the problem would require a major engineering change--and a lot of charm to convince developers to change their workflow.

"In the process of scaling, the company got a little too conservative, and by way of doing that started driving away some of the technologists and innovators," PayPal's CTO James Barrese says. "We are really going back to the roots of PayPal and reinventing the organization, the talent, the types of initiatives that we undertake, and the products that we build."

To speed up production cycles, PayPal converted a substantial slice of its data centers--a fifth so far--into a private Openstack cloud. Openstack is an open-source cloud operating system that controls the computation, storage, and networking resources within a datacenter. It allows companies to create their own public cloud, as an alternative to a service like Amazon Web Services, or to implement a private cloud within the corporate firewall for use by internal business teams. PayPal did the latter.

"We started in January and by the end of the summer we were already there, " says Barrese. "I would say that's a very accelerated timeline because we were on a mission to make it happen." Production cycles are now seven times faster than a year ago.

According to Barrese, the biggest challenges in the project were not technical--although those were considerable--but organizational. It's hard to cajole a technical team numbering in the thousands to work in a different way. Here's his advice on how to do it.

Avoid Analysis Paralysis

PayPal mandates 99.9999% uptime and must maintain stringent security and at the same time it needed to massively improve on the speed and flexibility of deployment. This makes selecting the right technologies difficult. "I would encourage people to just try it and not get into analysis paralysis, says Barrese. "A large organization can spend six months or a year trying to make 'the right decision' and at the end of the day there's no perfect answer.You are going to learn a lot more by doing it than you will by analyzing it."

This may mean working with immature technology. Openstack and other open source cloud technologies are still relatively young. PayPal has multiple data centers running in a mission critical environment, something which Openstack doesn't handle. "We are building where there are holes," says Barrese. "We have to."

Get Top-Down Support

While new tools and technologies often bubble up from development and operation teams themselves, Barrese says that for a project of this scale the entire organization has to be on board. "You have to have top-down support. We are putting the muscle behind making it work because it's not easy to go implement. If somebody is just trying to do it skunkworks in a back room, you can do a prototype but you are not going to do something at scale unless you have the backing of the organization."

Win Over Operations

Maintaining constant uptime using Openstack is a new and relatively untested process, but missing availability targets may not be the operation's biggest concern as provisioning becomes more automated. "There is a lot of unspoken fear where people are worried that it will eliminate their job," says Barrese. "What's critical is to get those people to see that this is the best thing possible for your career, because you are going to be one of those people who knows how to transform to a cloud infrastructure, and second those people are going to be able to work on much higher-value activities like proactive monitoring, advanced detection, being able to predict failures versus the reactive stuff of trying to diagnose a failure."

Stage And Scale

PayPal, like an large organization, had existing business targets and deadlines which still needed to be met while the technical overhaul was taking place. This meant that staging and sequencing was crucial. "The first part is you've got to get your band of pirates, the small team that's going to help you get initial installations done across engineering, quality, operations," says Barrese. "There are three steps. There's an initial proof of concept. Let's just say a low-risk business use case. If it has problems or it's delayed it's not the end of the world. The second one is an extremely high-volume use case that proves scale. The third I like is a spanning set of use cases which exercises a lot of functionality." Once you have proven that the new infrastructure stands up both at scale and in comprehensiveness, he says, you can start rolling it out to more and more teams.

Re-Engineer Engineering

Operations weren't the only technical teams in the front line. PayPal's engineers needed to rewrite their applications to run in a cloud infrastructure. Engineering had to start to build, test, and deploy their applications in a new way. "We are in the middle of wrapping up a really large transformation to agile," says Barrese. "We were doing things in old waterfall style. We threw all of that out the window. We have completed restructured our teams. We've co-located teams. We have incorporated DevOps."

The relationship between engineering and operations teams had to change. Engineering and operations usually have different goals and are rewarded for different things. "Typically what happens is that the ops team is measured by a certain set of production goals and the engineering team is rewarded for a different set of goals," explains Barrese, "I cross-wired and gave shared goals. The ops teams have delivery goals as well as availability goals and engineering teams have both delivery and availability goals."

PayPal now has one of the largest hybrid cloud deployments in the world and the company plans to eventually transfer all of its systems over to this kind of topology. Barrese's only regret is that he didn't start the project three months earlier. "You need to be prepared to get a little dirty in making this stuff work," he says. "It's still being built, but it's also an exciting time to make it real. The tectonic plates are shifting. Every company should be going after this."

This Smart Sleep Mask Helps You Rest Better in Less Time

$
0
0

Studies say up to a quarter of Americans don't get enough sleep, and there's no shortage of articles about how to function better with power naps and elaborate "polyphasic" sleep-and-wake cycles.

But without actually spending the night in a sleep lab, it's hard to actually measure how much and how well you're sleeping. Warsaw-based startup Intelclinic wants to change that with a smart sleep mask called the NeuroOn that will monitor eye movements and brain waves to track when users fall asleep, how much sleep they get, and how frequently they wake during the night.

Intelclinic's raised more than $200,000 so far through an ongoing Kickstarter effort to fund production and further development of the NeuroOn masks, and the company aims to start getting masks to Kickstarter backers in Spring 2014, said founder and CEO Kamil Adamczyk.

Users will be able to monitor their sleep schedules through an iPhone or Android app. The masks will automatically upload sleep data to smartphones through low-power Bluetooth connections when users take them off. And the NeuroOn masks can automatically wake users at a light-sleeping point in their sleep cycles so they don't get up groggy, Adamczyk said.

"Our mask monitors your sleep and also wakes you up very gently," he said. "You'll feel rested, and you're not anxious and so on."

The NeuroOn can help users looking to get more hours out of the day by maintaining a polyphasic sleep schedule--that is, by sleeping less during the night and supplementing with additional, shorter sleep phases (read: naps) throughout the day.

Adamczyk said he's using a beta version of the mask to help him maintain an "Everyman" sleep schedule, with about three hours of solid sleep during the night and a few naps during the day.

"I started to sleep polyphasically a year ago," he said. "At the beginning, I started without the mask, but right now I'm the first beta tester and I'm using it constantly."

Polyphasic sleep is controversial in sleep science circles, Adamczyk acknowledged, saying getting seven to nine hours of sleep per night is still believed to be the healthiest way to rest. But even for people getting a full night's sleep, the mask can still be helpful, tracking how long it takes to nod off and helping to wake users when they'll feel most refreshed, he said.

"We can detect at which phase of sleep you are at the moment, we can wake you up 10 minutes before you're alarm clock but at a very light stage of sleep," he said.

The company's also working on code to generate individualized travel sleep plans to minimize jet lag, he said.

"The point is to plan your sleep before and after the trip to minimize the jet lag effect for you," he said. "So it depends on the start and end point of your trip--we are creating time zone maps and preparing a sleep plan for you to minimize jet lag."

The mask works by using electronic sensors similar to those found in an electrocardiogram to measure eye muscle and brain activity, he said. In current prototypes, each mask has three sensors, and Interclinic's planning to upgrade to seven for greater precision, he said.

"Our brain generates electrical activity and we can measure those electrical activity by those electrodes," he said. "It's the same story with your heart rate; we're using the same technology to measure."

But the devices are designed to be comfortable enough to wear to bed.

"On the inside, we make the NeuroOn from soft, comfortable materials, with the ability to adjust to your face, thanks to the use of viscoelastic foams--the latest version of memory foam," the company wrote on its Kickstarter page.

Those sensors will help the masks work better than other smartphone sleep monitoring apps, which typically only measure movements by using phone accelerometers, Adamczyk said. Those measure sleep phases less precisely and can also be thrown off by other movements, especially if users share their beds with partners or pets, he said.

"If you're sleeping with another person in the bed, there is no option to measure your sleep process," he said. "We decided to create a device which would be much more precise than those kinds of devices."

The masks might also be able to help trigger lucid dreaming--a phenomenon where dreamers are aware that they're dreaming and can even take control of their dream selves. In theory, the masks will be able to detect when a user starts dreaming and shine an LED light bright enough to be visible and incorporated into the dream but not bright enough to actually wake the user up, Adamczyk said. But, he acknowledged, the very existence of lucid dreaming is, like polyphasic sleep, controversial among sleep researchers.

In the future, Adamczyk said later versions of the masks might be helpful to astronauts, who need to maintain steady sleep cycles in the absence of normal night-and-day signals. And, he hopes, in a few years, the masks--which he emphasized are not now intended to be medical devices--may be able to meet regulatory standards to be used to detect sleep disorders like sleep apnea.

"It's a long story, and I think it will take another few years to get this approval," he said.


Disabled iPhone Owners Are Desperate For A Jailbreak

$
0
0

Few people are as anxious to jailbreak iOS 7 as Chris Maury. It's not just the usual productivity hacks and customizations he's waiting for--you see, Maury is going blind, and the special accessibility tools he needs to use his iPhone are only available to jailbreakers.

To help speed the process along, Maury--who was diagnosed with a degenerative vision disorder three years ago--recently joined forces with a crowdfunding startup called Threshold to try and encourage developers to poke holes at Apple's fortress a bit more forcefully. The motivator: a cash prize, $7,300 to be exact.

As you might imagine, the plan has been met with some controversy. Will it work? And should it?

Why Jailbreaking Matters For The Visually Impaired

If past timelines are any indication, iOS users should probably expect to see an untethered jailbreak sometime in the next few months. Maury and Elizabeth Stark, founder of Threshold, realize this. For them, the issue is speed. For users to rely on jailbreak-only tools, the 3-5 month lag between new versions of iOS and an untethered jailbreak is just too much.

So why, some will be quick to ask, don't visually impaired users just switch to Android, where the sky's the limit when it comes to customizing the experience? Well, even though Android has improved in the accessibility department over the years, it's still not as good as iOS; Apple has long been well-regarded for its range of disability-friendly features in Mac OS and iOS. But for Maury, even the best is not good enough.

"There's a lot of functionality that the blind community gets out of the jailbreaks that you lose as soon as there's an update," says Maury. "And you're just not able to get access to those features until a jailbreak comes out."

The tool Maury relies on the most is f.lux, a Cydia tweak that lets users adjust the color tones on iOS and reduce eye strain for those with visual impairments. Other jailbreak-only features include the ability to install more accessible keyboards and alter the speed of Voice Over, iOS's native voice guidance tool for the visually impaired.

"The iPhone replaces a lot of stand-alone hardware devices that used to cost $300 or $600 and are now a $9 app," Maury says. "But because they have such tight control over the OS, there are certain things that you can't do." That is, without jailbreaking.

How A New Idea Sparked A Heated Debate

Last week, Maury and a handful of collaborators launched IsiOS7JailbrokenYet.com, a site soliciting donations for a reward to the first team of developers to come up with an untethered, open source jailbreak tool for the latest version of iOS.

"We're interested in incentivizing other people to address the problem and work on the problem," says Stark. "We ultimately think there are people that are capable of doing it and are actually working on it as a result of the fact that we put this out there. They might not have done it otherwise. Some of these people are very talented and maybe can do it sooner."

Backed by the likes of BoingBoing's Cory Doctorow and iFixIt CEO Kyle Wiens, it marries three popular concepts--crowdfunding, XPRIZE-style rewards and iOS jailbreaking--into an initiative one might expect to be a slam-dunk. Instead, it's left the jailbreaking community divided.

The most resounding objection to the plan has come from Cydia creator Jay Freeman--known online as @saurik--who took to Reddit last week to criticize the idea of offering financial incentives to encourage a jailbreak.

"The primary problem I have with this website is that it attempts to change the dynamics from one of 'people who do things that are fun to make devices more open' to one of 'people who do things to win cash prizes,'" Freeman wrote. "I've seen the effects of bounties in the Android ecosystem, and they are quite negative."

For Maury and Stark, Freeman's condemnation of the project is a bit of a PR headache. Few developers are as well-known and respected in the iOS jailbreaking community as the man who built the Cydia app store itself. Still, other developers have been more receptive to the idea. Comex, the former jailbreaker who was famously hired away by Apple after creating a tool called JailbreakMe, tweeted cautious support for the project, saying that financial motivation "might be necessary" to help speed the jailbreak process along. Meanwhile, Stefan Esser, a German researcher who developed an exploit for iOS 4.3, said he thinks the most prominent critics "fear to lose their monopoly." As is typically the case, the debate continues on Reddit.

"I think Freeman makes a lot of really good points and he's right that they do have the potential to change motivations," says Maury. "It allows people who doesn't necessarily have the skills but definitely have the need--like the disabled community--to incentivize and encourage solutions to the problems that they would otherwise not be able to."

Want To Get More Retweets? Try Doing This (And This, And This)

$
0
0

A new report from TrackMaven analyzing 1,423 Twitter accounts and well over a million tweets was released today, showing exactly what makes a successful tweet--that is, a tweet that is re-posted by a lot followers. Here's what the data say:

Time of day matters: Although peak Twitter hours coincide with the workday, retweets are actually higher during our off hours. On average, we retweet more on Sundays than we do on any other day of the week, and Twitter sees more retweet activity from 10-11 p.m. than it does for the rest of the day.

Hashtags and at-replies help: The use of hashtags and at-replies was also found to have a positive effect on the amount of times a tweet would be retweeted. Part of that is common sense--part of what makes both of them such useful tools is that they give your tweets a much higher visibility--but the study also found that there are diminishing returns on tweets with more than six @ mentions.

Don't neglect the period: However, when you do incorporate @ mentions, don't forget how valuable the period can be. Beginning your tweet with @ makes it visible only to those who follow both you and the user you mention. Beginning your tweet with a period when it's necessary to start with another user's handle makes sure that it will go out to all of your followers.

Also, the report reinforces the importance of using images to take advantage of Twitter's recent redesign. Now that Twitter feeds display in-line image previews, tweets with images are almost 25% more likely to get retweeted--an increase from .404 retweets/1K followers to .496 retweets.

Good news for the slew of Hulk-inspired parody accounts too: tweeting in all caps will net you .8 retweets on average per 1,000 followers, as opposed to tweets with only one-tenth of their text in caps, which averages out to .147 retweets per 1,000 followers.

So don't bother using your inside voices. Add some text...

How To Share Files, Snapchat-Style

$
0
0

On Wednesday, a company called Younity released the beta version of something called Ephemeral File Sharing--think Snapchat for files. Users can share files with a self-destruct timer to a list of friends, with options to share read-only versions. The default timer expires after seven days, but files can be unshared at any time.

Younity serves a larger purpose: uniting all your hard drives. It's a problem made widespread by the adoption of mobile devices: When working with platforms that obscure the file system, or aren't connected to the other file systems in your life (such as your computer), how do you go about sharing and keeping track of them? As we've written about before, Younity uses a patented "ubiquitous data protocol" to create a personal network that allows you to access and share files across devices without syncing. The goal: give you access to all your files on every device, everywhere, over the air--whether or not those devices are powered on.

Essential to Younity's approach is the file overload that can occur when data is used and shared across many devices. When you can't really dive into the file system of a particular device without significant workarounds, multiple copies of a file can stack up needlessly, and obsolete documents sit there taking up what is often very limited storage space. Younity's proprietary protocol exists primarily to trim bloat by removing a file's specific location from the equation entirely.

Philosophically speaking, the Younity approach is markedly different from Google Drive and Dropbox, which are incentivized to make you grow your data pile as fast and big as possible. But since Younity isn't a cloud data provider--just a file management architecture--the company is focused on letting users get more out of the limited space they have. Perhaps hard drive space will someday be so absurdly cheap that deleting files will feel like an anachronism--til then, we like Younity's solution.

Classify Social Data With DataSift's DIY Machine Learning

$
0
0

DataSift aggregates social data from sources like Twitter, Facebook, and blogs and lets you query it via a single API, but making effective use of that data is still a big problem for many businesses--until today.

"A lot of the trouble that you have when you try to integrate social data into an actual business," says DataSift's CTO and founder, Nick Halstead, "is what is the pivot point to connect the social data to the business behavior? So you need to know what the discussion actually means rather than just finding that somebody is talking about you."

Every business is different and therefore may use the same data in a different way. Are you an airline which wants to automatically classify new customer conversations as queries, complaints, or urgent requests? Or a B2B business which wants to determine the profession and seniority level of Twitter followers in order to identify leads? DataSift's professional services team often builds custom tools like Dell's Social Net Advocacy for this very reason. Companies with in-house data scientists might do the same using DataSift's data.

DataSift's new intelligence engine, Vedo, allows developers with no data science background to build custom models of customers and conversations and use them to automatically classify the real-time social data available via DataSift's API. Vedo can even build those models for you using Machine Learning.

A classification model automatically sorts new examples of an object like a tweet or Twitter profile into a number of classes. DataSift created 30 classification models and released the code so that they can be used by developers as examples. One of DataSift's sample models takes every new message sent to an airline's customer service account and categorizes it as a "query," "rant," or "rave" based on the words used in the tweet.

Another example distinguishes Twitter bots from humans. "Tweets always come from a source like TweetDeck," says Halstead. "There's about 80,000 of them. If you see a tweet coming from TweetDeck you know for sure that a human sent it. There's no way to fake that easily. TweetDeck users, on average, follow a daily 24-hour volume curve. We built a model to identify bots by knowing when they are tweeting at high volumes out of sync with that 24-hour cycle."

DataSift's query language, CSDL (Curated Stream Definition Language), is used to develop custom classification models. For example, you can create a rule like "tweets which contain the word 'help.'" You can then add a tag to the rule-like query which indicates that all matching tweets which match the rule are queries. Tweets which match a different rule might be classed as complaints.

The entire set of rules, which can be arbitrarily complex, form the classifier model. Each element of a rule may also have a score attached to it, e.g., "help" might get a score of +10, while "can" gets a score of +1 when classifying a query. The rule with the highest score indicates the class of the tweet. Once completed, the classification model is uploaded to DataSift, where it can then be used to classify new or historical social data.

Vedo also allows you to programmatically train and use a Bayesian classifier, a machine learning algorithm which automatically learns a classification model. Instead of building a set of rules manually, you take a set of tweets or whatever social data you want to classify, and manually attach a label-like query or complaint to each one.

These labeled examples are called a training set. Each tweet will have a number of features like the words contained in the tweet. A Bayesian classifier builds a probabilistic model of feature values using the training examples. If, for example, the word "help" appears much more frequently in queries than complaints, then the classifier will consider it more likely that a new unlabeled tweet is a query if it contains the word "help." The model the Bayesian classifier learns from the training set can be used to predict the class of new tweets based on their feature values. While the accuracy of a classification model built manually depends on the domain knowledge of the developer, that of a Bayesian classifier depends on how representative the training set is of the true population of tweets.

DataSift also takes care of the low-level text processing required prior to running Machine Learning. "You can't even start doing text processing in Chinese or Japanese until you have word-chunked the characters because Chinese and Japanese have no white spaces," says Halstead.

"We have spent the last year reading Japanese and Chinese, lexically parsing it, inserting fake spaces where the whitespace should be (between words) before we try and look for a word. You can't do Machine Learning against Chinese or Japanese unless you have have done the chunking."

DataSift, which just raised a new funding round of $42 million, sees analyzing enterprise social data as the next frontier. "In Q1, we will open up a public API where you can insert any unstructured data into DataSift," says Halstead.

"McKinsey says that 70% of the world's business data is unstructured. Instead of looking at public social data you can suck in your employee discussions and see what's going on inside your business. We already ingest Yammer data. Every day DataSift is working out out what topics people are discussing. What big pushes are going on? Are the employees really getting behind it?"

From Building Mac OS X To Building... A Vaporizer?

$
0
0

Hardware and software development are pretty different. But they clearly have some overlap and mutual appeal because these days an increasing number of programmers seem to be developing for the ultimate UX, the physical world. So we figured we'd find a good example--Mark Williams, a former Apple dev and Sasha Robinson, a former Juniper dev who partnered to try and build the best handheld vaporizer out there. We asked them about working on a software UX versus an object for the real world.

"Hardware pushes back much harder than software does, because the real world pushes back a lot harder," says Williams. "Achieving the same level of elegance of user experience is a bit harder than in software. And with software it's easier to achieve your optimal user experience because software has fewer limits."

On the flip side, Williams says, "software can be really abstract and esoteric, while hardware brings with it a whole lifetime of physical experiences that people already have. The ultimate goal, as I learned at Apple, is a user experience that integrates hardware and software."

How This Hardware Project Ignited

The Firefly took about four years to develop and started with the two working on concepts in Robinson's basement. Williams and Robinson say that they just didn't want to smoke anymore, and after meeting at Burning Man it seemed like their priorities were in line to try to best handheld vaporizers like the PAX or even tabletop models like the Volcano.

Robinson had burned out developing proprietary software for high-speed network boxes at Juniper, because he was working long hours on a product he knew no one would ever interact with directly. He spent four years doing welding and metal work before joining the design consulting firm MOTO (which was bought by Cisco in 2010). And Williams was ready for a change after five and a half years working on OS X for Apple. The two collaborated after hours on Firefly for about six months and then both left their jobs to fund and develop the product full time.

"In terms of where we met, it was through a Burning Man decompression party in San Francisco and then camping together at Burning Man," Robinson says. "And through a lot of talks about what we wanted to do and how we wanted to create a next stage in both of our careers. And we saw this product that we both wanted to have in our lives as an opportunity."

Williams and Robinson brought complementary skill sets to the development process. Williams has strong visual and spatial skills, so he worked on a lot of design aspects, while Robinson planned and executed much of the firmware that manages things like the heating element. But both are adamant that every detail, whether it related to the buttons on the device, the cleaning mechanism, or the mouthpiece, was ultimately a collaboration.

"Sasha and I developed together under a common vision," Williams says. "With the Firefly, we knew that the hardware and the body had to be super robust and strong to deliver the actual physical experience. But to do so we needed a brain-executing software in the background to give the digital control that we wanted, but still allow for the user to feel an analog experience."

Treading Carefully Around Regulations

As with any product that is marketed for (ahem) tobacco, Williams and Robinson had to work with import/export and tobacco lawyers to make sure they were complying with U.S. customs and import laws.

"You do have to spend a lot of money on lawyers just to have them tell you that you're doing something or shouldn't do something that's kind of obvious," Robinson says. But both know how high the stakes are for complying with regulations. "When you're importing stuff into the U.S. the jurisdiction is federal," Williams says. "So you have to be really sensitive and respectful of that."

In terms of designing the product for a user who might be in an altered mental state, Robinson's thoughts seemed to echo his view that an "operating system's job is to disappear." The goal with the Firefly's design was total accessibility involving minimal effort or practice.

"When you create a device that is as easy to use as possible and has as little decision making involved as possible, you've created a device that's accessible for anybody in whatever state they happen to be in," Robinson says. "What we focused on is the easiest, most approachable experience."

The Firefly has a bigger footprint and more weight than the PAX, which makes it feel substantial and almost luxurious. It's heavy for a portable, though, and wouldn't be very subtle in a public space. Testers also found that battery life wasn't very long, about enough for one smoke session--either the extra weight wasn't coming from a bigger battery, or the heating element draws a lot of power.

The entire front of the device is held in place with magnets, and you remove it to fill the bowl. It's a nice mechanism, but you have to be careful to align the front exactly when you put it back on because testers found that it didn't snap perfectly into place on its own, and using the device with this misalignment affected the vapor quality. The unit Co.Labs tested got pretty hot near the heating element, but not dangerously so. Sometimes there was a little bit of particulate in the vapor, but overall testers had the same reaction: "It's smooth."

"Smoothness comes from actually getting to enjoy all the flavors and active ingredients and not just a single one," Williams says. "That's why we built our heating architecture and our control mechanism around this system that ramps up the heat each time you inhale to 400 F. We digitally control it so it doesn't get beyond that, and so along the way when you're taking five- or 10-second inhalations, you're basically ramping up from room temperature to 400 and you're getting the vapor of everything along the way."

The Firefly retails for $270 and comes in three colors. Williams and Robinson say they want to eventually work on expanding the Firefly into a line of products after this launch. "We're actually super eager to get back into the workshop," Williams says. "That's what we really, really love to do."

Viewing all 36575 articles
Browse latest View live




Latest Images