Turning the Tables at a Messy Hackathon
At the TechCrunch Disrupt hackathon in New York, April 27, 2013.
There are two kinds of hackathons that I’m familiar with. The first, of which Evernote produces more than a dozen a year, are focused on one (or maybe two) platforms — like the Evernote API. Companies like ours will set up a space for a 24-hour coding spree, invite a bunch of developers, and parachute in a few engineers and evangelists to help out.
Focused hackathons are fun and productive. They’re a chance for us to bond with young (usually) developers, and to learn what these devs are looking for as they bang on our platform. We always get cool projects from these hackathons, like Context Booster from our recent Berlin hack, and Epicnote from one before that in Sao Paulo.
But I’m partial to the second kind of hackathon: the Messy Hack. These hacks, like the TechCrunch Disrupt weekend hackathons that usually precede the official Disrupt conferences, are massive and chaotic. They attract hundreds of developers — the last TechCrunch hack in New York had about 600 — and dozens of API provider companies. I was at this Messy Hack. It was glorious.
At these hackathons, platform companies like Evernote have to compete with other platform companies (In New York they included Twilio, Microsoft, Box, and others) for attention and love from developers.
I spent more than 20 years covering startups for the tech press, and I’m accustomed to watching new tech company CEOs pitch their hearts out to media, funding sources, and potential partner companies. The pecking order is clear: The startups are the supplicants; the established companies and media are the power brokers. I’m used to being in the power seat.
At a Messy Hack, the dynamic is reversed. The coders are in the audience, doling out attention to a string of platform evangelists who take the stage, one after the other, pleading in their brief pitches for the devs to code up apps using their APIs.
It’s a bracing reversal. Personally, I find it fun to pitch for a change, instead of being pitched. But it’s also a great opportunity to refine your business and your message. You can tell, after a few short hours, if your pitch is successful, and by extension if your company’s platform strategy is working. You tell by the number of developers who come up to your table during the hackathon to ask questions, and then you tell, at the end of the event, by the number and quality of the projects presented.
Another great thing about Messy Hacks: Mashups. With multiple APIs being supported at these events, developers are more likely to go off and build apps that bring different platforms together.
Don’t bother trying to leave. Just sit down and code.
Hacking the hackathon
Now, it’s not hard for platform companies to game these hackathons, by offering prizes to developers who use their APIs. We have done that, and it does drum up interest from hackers who come to have some fun and maybe win a few bucks over the weekend. In New York, some companies with serious APIs offered prizes, and they got some decent projects. But at least one company offered a prize that was too big for the value of its simple API (in my editorial judgement) and it got more projects created than it deserved. Hacks created just to win a prize have a lower probability of moving to the next stage of development; the payout is too much and too early.
At the recent New York TechCrunch hackathon, we offered no immediate prizes, and yet we were gratified to see six teams present interesting projects at the end of the event. My favorite was Squirrel, which is creating a nice-looking reading list that’s made up of the articles users clip to Evernote. We also liked the hacks Everslide, which makes quick presentation slides from Evernote notes; and Evernote-Quick Team Manager, which mashes Evernote together with Highrise or Basecamp for group task management.
It was all the sweeter to see these projects given the competition we were up against and the fact that our only prize was far in the future — well-developed Evernote integrations might win our three-month-long Devcup competition and some of those winners might be invited to our Evernote Accelerator. And, to be fair, there were 183 projects at the hackathon, so seeing six Evernote projects didn’t mean we “won” the event by any stretch.
But the developers who built for Evernote did so for the best reasons: They saw value in our API and our base of users. While we’re not above sweetening the pot a bit to encourage developers who are on the fence to to hack at Evernote, in this case, we didn’t even do that and we still got neat projects.
A competitive hackathon where you don’t give out prizes is a brutal opportunity. But the payoff is sweet.
-Rafe
Audio version of this column
Our next hackathon
Join Evernote and Honda at our first Design & Build Weekend at Evernote HQ!
Hold the Line
I was lucky. A week after I got in the line for Mailbox, it was 800,000 people long.
The Mailbox line concept is catching. The new Tempo calendar app has a line. The Web-based email service Mailstrom does, too.
So the line’s the thing, right? And now you’re thinking about implementing a line for your own new app.
I suggest that you seriously consider this option before proceeding. Think about what a Mailbox-like line means for your users, and what you’ll get and give up if you use a line instead of a more traditional beta signup process.
The biggest decision point is to make sure that throttling inbound new users is important, and not just show. Mail apps, in particular, are special: Each new user generates a huge spike in utilization as the service reads in their existing email. Nobody can realistically expect a startup to process a million GMail archives all at once. If that’s what you’re doing (or something similar, like reading in users’ Twitter feeds), you do need a way to keep your load manageable as you grow, and a system that lets users in one at a time might keep you from failing the moment you launch.
Welcome to Tempo. Please hold.
Other apps, where new users start as data newborns, with empty accounts, are much less likely to knock over a server on startup. So don’t fake it. If each new user only adds a touch of load, don’t pretend you’re moving heaven and earth for them.
Here are some more pros and cons of lines.
The big Con: You will get slammed in the App Store
People don’t like downloading an app only to be told they can’t use it. You’re going to annoy a lot of users. A lot. In a competitive market, your low, one-star rating might hurt.
Sort of a Pro: You control when your App Store ratings reset
When you release a major upgrade for your app, the App Store ratings start over. So once you’re done with the line, release a new version to wipe out all those one-star reviews.
Pro or Con, depending: It’s transparent
If you have a massively popular app, like Mailbox, it can be a big marketing coup to say that there are 800,000 people in line to get the app after only a few mentions in the tech blogs. It also looks good if you’re blasting through invites at several spots a second. But if your line gets stuck for days or moves slowly, and there are only a few thousand people in the line and the user doesn’t feel he or she is getting any closer to the front, it can dampen enthusiasm for the product.
Mailstrom uses a line for its Web-based service. That’s new.
Con: You can’t tell people how long the line will be
If you could tell new users when they’ll be admitted to your app, then you’d just tell them, right? The idea of using a software line is that you can slow it town, speed it up, or stop it altogether if you need to as your user base grows and you learn how your service handles the load. That means that there’s no way to give users a Disneyland-like estimate on their time to reach the end of the line. That can be frustrating.
Pro: Believe it or not, you can get away with this now
Before Mailbox, there was no way a small developer could have gotten a non-functional app through Apple. And Mailbox, for users coming in now, is just that: It’s a number on a screen that says, “You’re late, sucker.” But the team at Mailbox, I hear, worked their connections with Apple to get approval for this new type of in-app velvet rope. So now you can do the same thing, too.
But only do it if you desperately need to, and be aware of the numerous ramifications of this concept.
Don’t add a line to you app just to look cool. You’re not running a nightclub. But if you have a mobile app that you expect will have scaling issues, then a line can be a good way to throttle users and keep them informed at the same time. Just don’t forget to update your app to reset those awful ratings you’re going to get.
-Rafe
Further Reading
Mailbox Stops the Email Bleeding (Opportunity Notes)
The Thinking Behind Mailbox’s 800,000-Person Waiting List (Fast Company)
How Long Is The Mailbox Line? (Waxy)
Flipboard’s service collapses on iPhone app launch (ComputerWorld, 2011)
LaunchRock sets up site launch placeholders (CNET News, 2011)
GMail only Shed its “Beta” Tag After Five Years (NY Times, 2009)
Audio version of this column
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
Mailbox Stops the Email Bleeding
Mailbox makes archiving easy: Just swipe on a message header. The green bar and checkmark appears. Release to archive. To delete, swipe farther.
I have replaced the default email client on my iPhone with a new app that does less, and it’s made me a happier person.
The new app is — no surprise — Mailbox, the darling of the tech press (see Further Reading, below). Mailbox is perhaps the first iPhone email app to match the old Blackberry email app as workable product for doing email triage on a tiny device on which you steal moments when you’re bored in a meeting.
I’m not saying other other email apps don’t let you do this, but none seem to be as dialed in to the mobile experience as Mailbox. None seem to say, as Mailbox does, This is where you stop the bleeding. For major surgery, fire up your real computer.
What are the elements of triage, as it relates to email? Mailbox has broken it down into three elements.
1. Deleting
Probably the most productive thing you can do on a smartphone email client is clear out the garbage. Not only should it be easy (which it is, in Mailbox), but spiking messages is just about the most intellectual activity a distracted mobile user can be expected to do. Mailbox has a simple and fast way to archive or delete messages. It’s one press faster than the iPhone’s native delete function, and that makes a difference.
2. Shunting
Mailbox shares some genetics with to-do apps. One key activity in the product is the capability to boomerang messages back to you for later (see also the unrelated GMail plug-in, Boomerang). Know you’ll want to answer a message when you’re at your desk? One swipe and a tap, and it’s gone until the time you specify (like “This Evening,” or “Tomorrow,” or “In a month.”) With this feature, you can stack up your email to-dos for the times you think you’ll be able to deal with them. It lets you control your email panic. And you won’t lose anything. As Mailbox creator Gentry Underwood told me,” Everthing will come back. Just not all at once.” (Feature request: How about an option for “After my next meeting,” since the app should have access to my phone’s calendar?)
To get a message off your plate until you want it to bother you again, you can send it to your future self.
Mailbox also lets you direct messages to a few different lists, like “To read,” and “To buy.” You can also set up your own lists.
Again, it’s all about getting the patients out of the ER — your inbox — and into the clinics they should be treated in.
3. Speed
Under the hood, Mailbox is not a full email client like most apps. It doesn’t connect to email servers and initiate the IMAP protocols to retrieve mail. That would be slow. Instead, Mailbox on a phone communicates only with Mailbox’s own servers, and those servers contact the email machines (Google’s only, so far) on users’ behalves. The Mailbox protocols are fast and compressed.
The downside is that Mailbox’s own infrastructure is a single point of failure and a potential bottleneck. That’s a risk Underwood told me he’s willing to take (in fact, Mailbox has already had at least one outage).
The Opera Mini browser, by the way, works the same way: Opera’s servers retrieve Web content for the Opera Mini app and then send it, compressed and formatted for mobiles, direct to the devices.
These three main concepts work for Mailbox and its users. Mailbox doesn’t pretend to be a desktop app. You cannot do a lot of heavy organization in the app. You can’t set up complex filters. You won’t be writing super-long emails. But you can get email busywork handled on your smartphone more quickly, leaving the heavy lifting for a time when you can comfortably settle into your office chair and really focus.
So what are the elements of triage in your mobile app? What can you do to make your app as fast as possible for the user who only has 20 seconds to spend in it? How can you offload work from your mobile app to a server somewhere else or to your user in another setting?
A smartphone may be as powerful as a full desktop computer, but your users, when they are engaged with a mobile app, are much less capable than they would be on a proper computer. They have less focus, less time, and less patience. There’s no shame in creating an app that respects their diminished capacity when they’re mobile. As long as you also let them dig in to real work when they are not.
-Rafe
Further Reading
Mailbox: Swipe Your Way to a Clutter-Free Inbox (All Things D)
Mailbox Review: Your Bad Email Redemption (Gizmodo)
Is Mailbox the Best Email App for iPhone Yet? (Mashable)
Mailbox for iPhone: a next-generation email app inspired by Sparrow and Clear (The Verge)
There May Still Be Life in the Email Business (Opportunity Notes)
Audio version of this column
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
The Accidental Athlete
The Moves app tracks your walking, running, cycling and driving — all the time.
I own a Fitbit, but I rarely use it. When I go out for a run (don’t laugh, it happens), I have instead been using the RunKeeper app on my iPhone — because I never leave the house without my phone, and it does a good-enough job of collecting fitness data. But sometimes I do forget to turn it on.
Fortunately, there’s a new iPhone app, Moves, that constantly and persistently records my activity. It knows if I’m walking, running, cycling, or when I’m in something motorized, and it always knows where I am. It creates a record of everywhere I go, the routes I take, and by what means I travel. It tracks steps and distance. It is always with me since my phone is always with me, and more importantly it’s always on. It does impact battery life, but I can still easily make it through a day on my iPhone. I think it’s one of the coolest apps ever.
And it makes me worry about the fitness gadget market. Now, for people who are actually serious about fitness, a dedicated gadget is going to be lighter than a phone and might also provide better data. Likewise, a dedicated fitness smartphone app does more than Moves does with data — although you do have to remember to launch it and then to turn it off when you’re done, so it doesn’t devour your battery.
But for someone who’s not too terribly serious about health, Moves, and apps like it, are the future.
Sleep Cycle tells me it was not a restful night. But at least it woke me gently.
Here’s another example: Sleep Cycle.This app measures the quality of your sleep based on how much you move around in bed. You run the app, place the phone on your bed, and it gets your activity data from the phone’s accelerometer. It competes with the Lark, and other multi-use movement trackers (like the Fitbit again), and Sleep Cycle is limited in obvious ways compared to devices that you wear, but if you’re curious about your sleep patterns (and you sleep alone), it’s a great start. As a bonus, it’s also an alarm clock that wakes you up when you’re in a shallow phase of sleep, instead of potentially jolting you out a deep slumber as a non-aware alarm clock will.
These casual trackers could be great gateways to more serious products. Or, more likely, threats to an industry that’s just getting started. Several people in the health space told me the serious athlete or body measurement nut will go for a more accurate, wearable device. But I fear this perspective may be wishful.
It is true that a Lark, Up, Fitbit, Nike FuelBand, or Leikr watch will collect better data than an app like Moves, which actually requires a cloud server farm to process raw data from phones in order to accurately report on activity. But casual, persistent, phone-based activity trackers are going to get better. And they have the advantage of being automatically integrated into their users’ lives. The data they collect can correlate with GPS data, with calendars, and even social network data. Just as the cameras in smartphones are more-often used than dedicated cameras even when a standalone camera would take a better picture, the persistent, always-aware body tracking app is likely to put serious hurt on the low end of the emerging body-tracking gadget market.
I’m more interested in the upside, though. The idea that your phone can serve as a persistent monitor to your body is fascinating. It’s easy to poke fun at or be scared of a device that literally knows what you’re doing all the time, but there’s a ton of opportunity in analyzing this data for users’ benefit. I am eager to see where this market goes.
-Rafe
Audio version of this column
See me at SXSWedu!
Come to my panel, Raising the Modern Geek, with Stephan Turnipseed, president of LEGOEducation
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
Our Impatient Youth: iPad Game Measures Non-Academic Learning
Leo’s Pad features really good animation as well as games that measure more than they appear to.
For the most part, I am not a fan of “educational” smartphone and iPad apps for kids. As I’ve written before, many apps that claim to be educational are no more than poorly gamified basic lessons, and they fail both as education and as games. But it’s not really the fault of the medium. There are plenty of bad printed books, too.
It is a special, multidisciplinary challenge to successfully blend media, game mechanics, and an educational payload. It’s difficult to make educational apps appealing to kids and useful to parents and educators. There aren’t many apps that manage. But Kidaptive’s new Leo’s Pad app for pre-schoolers might be one.
On the kid side, you have an app that features, at least in episode One, a well-drawn and really nicely animated story, with little learning and assessment games sprinkled throughout. The few game elements I saw were a cut above the usual phoned-in game lessons I’ve seen in other apps. The games also adjust based on the child’s abilities. For example, a shape-fitting game will start off with very basic squares and adjust up to more complex shapes as the kid progresses and comes back again.
But what I really like about this app is that what it appears to teach, and what it really measures, are somewhat different. While the kid may be calculating how many rocks to put into an on-screen catapult to launch the character the right distance, the app is measuring not just arithmetic skills, but also non-academic skills like impulse control, patience, and fine motor skills. Results for these assessments will appear on a Parent’s Dashboard section of the app, coming in a future update.
The upcoming Dashboard for parents will show how your kid is progressing in non-academic areas.
Our schools mostly teach and measure academics. Parents do get report cards that include more than testing results, but usually not in a quantifiable way. So what’s a parent to do? There is more to achieving in life than learning book skills, and there are some things that parents or outside coaches are better at teaching than schools.
We may complain about our grown-up, online behaviors being over-measured, quantified, and data-mined for marketing and other capitalistic purposes. But analysis tools can also help us understand and improve ourselves, or something that we care about even more: Our children.
-Rafe
See me at SXSWedu!
Come to my panel, Raising the Modern Geek, with Stephan Turnipseed, president of LEGOEducation
Audio version of this column
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
There May Still Be Life in the Email Business
Everyone could use a better email client. But the business is a heart-breaker. Get used to a cool, independent app like Sparrow, and some giant megacorp like Google buys it out from under you, and takes its developers off the table. Commit to an open-source app like Thunderbird, and the foundation behind the app will begin to give up on it and orphan you, too. Even Thunderbird’s commercial cousin, Postbox, appears ready to move on.
Yet the optimism of developers springs eternal, or maybe delusional, because they keep writing email apps. There is, still, strong innovation in this market. I’m not sure email is a business, but that’s not stopping the devs.
Inky can sort emails by importance. It also auto-files social network messages, receipts, newsletters, and other message types.
Case in point: David Baggett, the driving force behind the slightly odd Windows and Mac email client, Inky. He’s working a two-fold plan with this app. First, the Inky email app is different. Its user interface is deceptively light and simple. Visually, it’s reminiscent of the Windows 8 mail client. But Inky does very aggressive automated email management on behalf of the user: It filters, it files, it ranks messages by importance.
To use Inky you have to learn to trust its sorting and filtering. The most recent messages, for example, no longer appear at the top of your inbox. Rather, the most important do. Mostly. The algorithms are still being worked out. Which is a big technology bet.
The second part of his plan is tackling the challenge of the email business itself. The outcome for an email product is far from certain, and the business model of Inky is built with that in mind. While Inky looks like a bare-bones email client, Baggett is not a “lean startup” devotee. Building the foundation of an email company is “at odds with the minimum viable product ideal,” Bagget told me, as an email platform is hard to create (news to me) and has intrinsic value that might not pay back for years. Baggett co-founded the travel data company ITA in 1997; it wasn’t until 2001 that the company got the Orbitz deal that made it a going concern; and it was another nine years before Google bought the company, for $700 million.
So the Inky client app may end up as just a demo front-end on top of a new smart email platform. It’ll take some time for businesses to see the value in that platform, but that’s probably where the money is.
Or not. If Baggett can get people and individuals to buy Inky itself, he’s happy to run the business that way. As he told me, “I gravitate to things where I have a good Plan B, where there are always multiple paths to value.”
-Rafe
Further reading
We need e-mail clients (CNET)
See also
Airmail, a great new Mac email client in the tradition of Sparrow
Alto, AOL’s new smart e-mail client
Mailbox, the stunning mobile email client, still in private beta
Audio version of this column
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
Get Carrot. Do it For the Kitten.
Carrot’s launch screen for me this morning. A family and sleep is no excuse.
You want a to-do list manager that will actually help you do things? Forget choosing an app based on features, or design, or some grand productivity philosophy. Get Carrot, the sadistic little app that will goad you to mark things off your list. It’s $0.99 in the iOS App Store.
Carrot has been getting good press for the way it brings cleverness and snark to the earnest list-management space. If you don’t get enough stuff done from your list, the app turns from light and airy to dark and brooding, and it starts to taunt you. Get stuff done, and it pats you on the back with encouraging but sarcastic comments, and other rewards, like trivia. As a to-list app, it’s very clean, if quite basic.
Carrot’s personality is part Siri, part HAL 9000, and mostly GlaDOS. She’s very funny. But there’s more to Carrot then her acerbic personality. There is, buried in this app, an entire narrative.
Carrot was created by Brian Mueller, who was about to start film school when he was waylaid by the need to help support his new family. He diverted his career into medical writing. But he never gave up on his desire to entertain. Meanwhile, he found that the to-do list managers he was using were too busy and too boring.
He learned to write apps, and started work on a to-do list manager for himself. He wanted it to be a game.
But Mueller noticed that while “gamification” had started to infect all kinds of apps, he thought that most gamified apps were paradoxically un-fun. “Worrying about levels and points becomes a chore,” he told me. “And it’s making developers lazy.”
Finally, Carrot is happy with me. More or less. I can tell because her screen is white, and she’s not torturing me by showing my tasks in Comic Sans.
Games, to Mueller’s mind, aren’t about points. “When I play games, it’s for the characters, writing, and personality,” he said.
That thinking begat Carrot. “Her” character and personality is just the start of the app. As you use it, Carrot reveals a story involving an orphaned kitten living in a data center, and an emerging, twisted AI (Carrot herself). The game and the narrative in Carrot are not just tacked-on in the form of points and gadgets. The app changes as you satisfy or disappoint it: You may find Carrot giving you nicer task-completion icons as you please her. Slack off and she may appear to lose your tasks to punish you.
If you’re of a certain mind, it’s actually quite motivating. It’s funny, and emotionally honest.
That’s because, I believe, that Mueller is an artist who knows how to tell a story. And by “artist,” I mean a person who has created a piece of work based on his own internal aesthetic, not for the masses and not for a focus group.
Mueller, though, is a vengeful creator. “When I started working on Carrot, I was worried that it might turn people off. But then I said, screw it, and went with it. The threat is the hook. It’s almost impossible to not make her angry. It’s for a certain type of person, a geeky user who’s more accepting of a sadistic AI.”
The story arc in Carrot does, sadly, come to an end. Mueller told me how. I wish he hadn’t, and I won’t spoil it for you. But Mueller does plan to keep the story alive. He’s having fun weaving the narrative and plans to keep extending the story to keep ahead of his user base, some 2,000 strong in the days after launch, that are barreling towards Carrot’s current conclusion. Even if he doesn’t keep writing, the to-do list function will keep working, since Carrot has several hundred random “rewards” and “punishments” not tied to the narrative.
I can’t say I recommend app-as-story for other developers, but in a brutally crowded app market, there is a lesson to be learned from Mueller’s philosophy: Don’t dismiss the idea of putting real personality into your app. You may turn off a lot of potential users, but the ones who do find their way to your product could end more strongly attached to it. Perversely so, in fact.
-Rafe
See also
Audio version of this column
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
3 Ways to Get Videos of the Kids to Grandma
“Send me the movie,” my mother says when I tell her I just took an adorable iPhone video of her six-year-old grandson. If only it was easy. Photos, sure: You can email them straight from the iPhone’s library. Videos? No such luck. You can share tiny snippets from videos via email or MMS, or upload vids from your phone to YouTube and share from there, but if you want to share a full iPhone video directly, Apple doesn’t (yet) make it easy.
Joya lets you create a list of people who will get whatever videos you add to your “stream.”
This missing capability is an opportunity that entrepreneurs are pursuing. If you’ve got grandparents in your life and want to keep them happy, or just want an easier way to privately share videos with a few people, without relying on a large public service like YouTube, here are three new apps you might want to check out.
Burst is the go-to app if you’re aiming to share vids of kids playing sports. The app has a cool editing feature that zooms and slows video playback when you tap an on-screen moment, like your kid scoring a goal. Burst also makes it very easy to add individuals to the list of those who will receive a video, and it keeps track of who you’ve sent to so sending them future videos is very easy.
Cloudee, from Boxee, is simpler. You can take a video with the app, or select one from your phone’s library, and when you add it to a “collection” in the app, the people you’ve subscribed to the collection will get an update notification, and be able to watch the video. I found the list management confusing at first, but I did successfully share a few videos with the service. It made my mom happy.
The new Joya is a a very early-stage app with an incomplete feature set, but it’s worth a shot as it’s laser-focused on sharing videos quickly with a small, mostly unchanging network of close contacts. The app is so focused on sharing that it doesn’t even have a link to the iPhone’s camera app; the founders say that most family videos are captured from the lock screen, the fastest way into the video recorder. But when it comes time to share, they say, there is no faster way to select a video and get it to your family or friends than their app.
There are, of course, other solutions for sharing videos. If you use a sync service like Dropbox, you can send links from there, for example. But I appreciate the focus on speed and usability (not completely successful, but improving) from these three family-focused, video-specific sharing apps.
I’ll add that it’s been a big mystery to me why it’s taken so long for apps that do this to appear. Perhaps developers have been holding off, expecting Apple itself to offer this function — which it should have by now. Joya’s Michal Bortnik also told me that the costs of providing a video sharing service are falling fast. “There’s a price war going on, and if you get to real scale, then it’s another order of magnitude lower.” So nobody has to shell out millions to build another YouTube.
It’s a pretty safe bet that, eventually, Apple will offer this service. But why wait? These apps provide this missing feature today.
-Rafe
See also:
Perch: Surveillance with a Heart
Socialcam
Viddy
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
Kannuu Makes TV Remotes Less Awful
“It feels like I write this headline every year or two: A new company has engineered a new and better way to enter text.”
I wrote that in 1997. It was the introduction to a story about Kannuu, which made a clever technology that used a four-way controller to quickly enter text on mobile devices. I liked it, but I also predicted that, “It probably won’t stick around long.” Sadly, I was right. The Kannuu four-way keyboard never got traction on mobiles. The current CEO of Kannuu, Todd Viegut, blames the global financial crisis of 2008 (he’s so familiar with it that he just calls it the “GFC” in conversation), and that’s certainly part of it. The rise of smartphones, with their on-screen keyboards, is the other.
But I was also wrong about Kannuu, because the technology has survived, and may well be coming back. This time, though, it’s landing on TV remotes.
Kannuu uses the four-way controller on your remote, plus predictive suggestions, to replace the annoying letter grids we’re now using on smart TV apps.
Kannuu is a predictive text-entry system. For each character you want to enter, you’re given four letter choices, one each on up, down, left, and right. Plus a fifth for “more options” in the center. As you “type,” the options change based on the most likely words or names you’re trying to find. You can’t exactly touch-type with Kannuu, since the mapping of buttons to characters keep changing, but neither do you need to actually look at the controller to enter text, and it is much quicker to use than the standard full-alphabet menus that most people are familiar with from on-screen search fields in TV-based applications.
It’s a much better idea for TV remotes than it was for smartphones, because you can operate a four-way controller by touch while keeping your eyes on the big screen. The little QWERTY keyboards that have popped up on some remotes don’t make that easy. The Kannuu system relies on a constrained dictionary of potential word hits (shows, actors, etc.); it wouldn’t be as successful as a general keyboard used for messaging.
The Australian company, resurrected from receivership after the GFC by Viegut, who specializes in this maneuver, is announcing partnerships this week with cable TV operators and TV manufacturers. That’s farther than it ever got in the cellular business.
The system is faster to use than letter grid menus. (Data and graphic from Kannuu.)
My prediction this time: It’s going to work, at least until tablet- and smartphone-based “second screen” entertainment guides start to replace remotes, or voice search replaces remote keyboards. I think both of these advancements are inevitable, but not for a few years, so Kannuu should be able to have a nice run while it figures out its next move. It’s all about staying just a few cycles ahead of consumer technology, and keeping moving.
-Rafe
See also
The Archos TV Touch Remote (Gdgt)
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.
Sync Different: When To Go Direct
I dig sync. I like products that make a user’s data available no matter where he or she is, even if it’s offline. My belief that synchronization is core to the success of information products is one of the big reasons I joined Evernote.
The cleanest way to offer data sync between devices is to have products replicate new data to a data center in the cloud, and then have the data center distribute the data (or deltas) down to users’ devices as needed. Syncing through the cloud is expensive, though: It requires the vendor pay for both storage and bandwidth. It is a major business commitment.
There is another way, without the expensive, recurring overhead: Direct, or peer-to-peer sync. But it’s challenging to get it right. So challenging, in fact, that it may not work to offer it instead of cloud sync.
Cubby syncs files directly, without using cloud storage.
I should note that at Evernote, we use cloud sync, and not direct sync, because we do a lot of processing on data in the cloud that is then synced down to users’ machines: Search across notebooks, image recognition, OCR, and our API for third-party integrations run on our servers. Cloud sync also means that the user doesn’t have to worry about keeping two machines on and connected at the same time when he or she wants to sync. On mobile devices, which may be connected intermittently, you can’t rely on any two machines being online at the same time. The cloud is always on. It makes for a more consistent experience.
Still, if you’re building a data app and want to offer sync as a feature (and you should), there may be a way you can take advantage of some of the benefits of direct sync. Here are a few companies that have done so, the most recent first:
LogMeIn Cubby
Cubby is a file sync and store service that puts its data in the cloud, like its competitors do. But the system also gives users the option to sync and shares files directly, device-to-device. Files synced or shared using “DirectSync” don’t count against the user’s storage caps, so users can share even gigantic media files over the system. If the devices are on the same network, data can also transfer at full speed, and not get throttled by a shared or slow Internet connection. LogMeIn sees DirectSync as such a big advantage that it’s actually a premium, paid feature (along with increase cloud storage caps), even though DirectSync files in transit don’t touch LogMeIn infrastructure and don’t incur costs for the company.
CrashPlan
CrashPlan is a backup service that offers device-to-device transfer at no cost, on top of the free client apps. If you want to use your mom’s computer as your backup destination, CrashPlan is completely free to you. When you want to use CrashPlan’s cloud servers to store your backup files, then you pay. The client app is the same regardless. In this company’s case, the free device-to-device service, in addition to being very cool, is great marketing material for the paid product.
Things
This is a task manager and to-do list app. The first version of Things offered only device-to-device synchronization of items. Unfortunately, it only synced between devices when both were running the app and on the same local network. (Other direct sync products use cloud services for signaling and coordination, even if the data itself transits the Internet.) This kind of sync required too much attention from the user. The current versions of the app now sync to the cloud.
FolderShare
In the old days, before cloud storage companies existed, a little product called FolderShare provided computer-to-computer file sync. For geeks who could remain aware that their computers wouldn’t sync unless more than one was on at a time, it was a great product. Microsoft bought FolderShare publisher ByteTaxi in 2005 and over the years has reduced the presence of PC-to-PC sync capabilities of its online storage products. In the most recent version, SkyDrive, the direct sync feature has been removed entirely. Microsoft’s Mike Torres said, earlier this year, “We won’t be offering a way to sync between PCs without using the cloud as we’ve found that it adds complexity. Both technical complexity as well as, more importantly, complexity in the overall user model.”
Direct sync is less expensive than cloud sync, but is definitely not just the cheap option. It has to be deeply baked into a product both technically and on the business and marketing sides, to overcome challenges of usability and utility. It can be done. The two products I cover here that offer direct sync, Cubby and CrashPlan, both exploit direct sync’s unique virtues to make it a value-add to products that are primarily cloud sync engines.
-Rafe
See also:
AeroFS
SparkleShare
A message from Evernote
Build knowledge. Share ideas. Get things done. Evernote Business.