In a few short months, Microsoft's Kinect has become one of the most exciting platforms around. Dozens of hackers are making use of the groundbreaking motion sensor, crafting projects ranging from quirky instruments to medical equipment replacements to art installations. Those thrilling projects all have one thing in common: Microsoft has nothing to do with them, and regular consumers have no access to them. You can't buy them in stores. And what you can buy in stores is disappointing at best.
For almost all of the ten-million-plus Kinect owners, including myself, the Kinect sits on the TV stand, collecting dust in between increasingly infrequent games of Dance Central--a launch title. Why is Microsoft letting their most exciting product in years--maybe ever--sit fallow?
I've been following the Kinect so closely because it has the potential to fulfill so many of our science-fiction fantasies. The Kinect is one of those rare gadgets, like the iPhone, that takes what came before--be it the motion-sensing trend started by the Nintendo Wii or early smartphones from Palm and BlackBerry--and moves it a step and a half further. The gadget and the ways you use it make sense, but it still feels like it was teleported here from three years in the future. People respond to these kinds of products: The possibilities they open up are so exciting, so new, and so obvious that ambitious developers can't wait to start exploring them. But for the new device to survive, it needs encouragement from its maker--and Microsoft is thoroughly dropping the ball on that front.
Some 99.9 percent of the world's ten million Kinects are hooked up to an Xbox 360, ready to play games. But where are they? The Kinect was unveiled in June 2009, nearly two years ago--Microsoft should have been able to provide developers with developer kits many months ago.
Microsoft has released about a dozen games after the first crop, but the majority of those are not exclusive to Kinect, meaning they generally just cram a couple of half-assed motion gestures into an existing game. They're also very low-profile games--several are cheap, licensed fitness games tied to such legendary wells of gaming creativity as the Spice Girls and The Biggest Loser. Ask someone with a Kinect what they do with it, and they'll tell you that they play Dance Central for those friends who haven't yet seen it, and that's about it. That's a serious problem.
What we've seen of the next round of Kinect games is promising. There's Child of Eden, the trailer of which is above. I'd be surprised if there's anyone who can tell me what's going on there besides someone frantically gesticulating at some sort of techno visualizer, but I definitely want to play something so unusual-looking. There's Rise of Nightmares, some kind of horror game from Sega that's only been shown in a trailer that features no gameplay (but a lot of screaming!). There are fighting games (this one uses your Xbox Live avatar as a fighter, so you could probably use the game as a way to take out your anger on your roommate for using your shampoo all the time), sports games (which could be fun, but which don't look to make much more use of the Kinect than versions for PlayStation Move or Nintendo Wii), and, um, hardcore marionette shooters. There's some sort of Sesame Street game designed for parent-child gameplay, which would be entirely uninteresting to us childless folk were it not for the fact that this game is helmed by gaming visionary Tim Schafer's Double Fine Productions, responsible for some of the most imaginative games of all time. There are ports from some unlikely places: a port of the popular iPhone/Android game Doodle Jump, which you will presumably control by tilting your body, and a port of indie darling Octodad, which you control through a series of arm movements as silly as the titular besuited octopus father--a very good thing.
These games look good, by modern video game standards, which is to say that many of them look utterly predictable and boring, and a few of them look like a hell of a lot of fun. But some of these are a long ways off, and many, like Rise of Nightmares, don't even have a proposed release quarter yet. All of the games I named are third-party titles (meaning, made by companies other than Microsoft), but the big MS still has the ultimate say in the release of games. Microsoft assures me that there will be some exciting stuff announced at E3 this June--but E3 announcements are not like Apple announcements, where Steve Jobs says "here's our newest product, you can buy it now." Usually, when a company announces a game at E3, it's destined to be released during the fall holiday season, several months later. Assuming Microsoft does the same thing, that'll make the Kinect's entire first year a total wash, without anything worth getting excited about. So I'm left to ask one question: Microsoft, where are our games?
Microsoft doesn't want us to think of the Kinect as just a videogame accessory, something like a hands-free PlayStation Move or Nintendo Wii. If you ask Microsoft, they'll say that's why they've enabled Kinect navigation for all kinds of other things the Xbox 360 can do--movies, music, sports, and television. What they won't say, but I will, is that every one of those implementations is totally unfulfilling. With Zune, Microsoft's music and video store, you can use the Kinect to control your media, but only to a very limited extent. Pretty much all you can do right now is use voice or gestures to perform extremely simple commands on the Now Playing screen--play, pause, scrub forward and back. It's nice enough, but it's all you get. The Kinect's voice command ability is amazing; why can't I say "Xbox! Play album 'No Color' by artist 'The Dodos'"?
Netflix for Kinect is the most recent, and one of the most egregious, missteps of this sort. The Xbox 360 has one of the better Netflix implementations out there, and combining Netflix with the Kinect seems like an obvious move. This is another example in which voice commands are preferable to gestures--it's more exact, and actually faster than using a controller, unlike gestures--but Microsoft again dropped the ball. The Netflix implementation is pretty much identical to the one for Zune, limited to basic commands while a video is playing. Even worse, you can't use Kinect to select even the movies that are in your own Instant Queue. The only videos you can select with Kinect are a few recommendations, which are often totally misguided, and, more importantly, are by necessity movies you haven't chosen to watch. Being able to pause your movie without a remote is great--but which do you do more often, pause a movie or browse through Netflix's huge collection? It seems like any one of the dozens of rogue Kinect hackers could throw together a superior version of Kinect-controlled Netflix in about ten days. Why can't Microsoft do it right in five months?
At least Netflix integration is here (and I can't really deny that saying "Xbox! Pause!" is pretty cool). What about the abundance of other features that Microsoft demonstrated and promised back when the Kinect was called Project Natal, and that have just...not been mentioned again. Take a look:
What happened to voice interaction with on-screen characters, like being able to say "Sure, let's fight!" to a scary vaguely Asian fighter? Or using your own gear in games? The idea of scanning your own art to be used on a virtual skateboard deck, or scanning any round object to use as a steering wheel in a racing game, had so many possibilities. It seemed to totally show up the Move and Wii: Those systems need you to clip their motion-sensing controllers into a variety of other plastic shells--guns, sports equipment, steering wheels--while the Kinect could just use anything you have lying around.
Some of those demos don't seem particularly feasible, which is fine. Virtual shopping has never worked very well, no matter who tries it, for example. But some of this stuff is eminently possible, if only Microsoft actually busted ass to get it out there.
Considering it took years of brilliant engineering and groundbreaking user interface design to release the Kinect, the one smart thing Microsoft's done in the five months since the Kinect's release is to recognize that the Kinect hackers are their friends. Rather than trying to box them out, Microsoft announced that it would release an official SDK--a technical toolkit that enables developers to access the Kinect's sensors and data to make their own projects--in May.
We spoke to Dr. Oliver Kreylos, who's responsible for some of our favorite Kinect hacks (like 3-D holographic video chat and 3-D video capture), about the challenge of developing for Kinect, and what Microsoft has to do to ensure a thriving app ecosystem. "In general," says Kreylos, "I'd say that developing the 3D graphics and gesture/interaction technologies that are required to create a compelling application are hard, and just adding a Kinect won't make them much harder." Developing for the Kinect is going to be difficult no matter what, at least in the early part of the Kinect's life. But there are lots of things Microsoft can do to make it easier--and Microsoft should do all of them.
The very obvious correct move after releasing the SDK is to create a Kinect app store. Developers are so thoroughly on board with the Kinect that they started developing within hours of the Kinect's release. Users love the Kinect, but given Microsoft's infuriating lack of follow-up, are getting bored with it. Microsoft absolutely needs to open an app store for the Kinect, and allow users to download apps to make use of their amazing new hardware.
Kreylos compares his predictions for the future of Kinect development to that of, surprise, the iPhone. When developers want to develop for a multitouch system like the iPhone's, they can either use a basic, free toolkit, which gives basic results; or they can expand on that on their own initiative to get more complex and interesting results (which requires more work); or they can use a third-party toolkit which has more advanced built-in functionality, but which may cost money. Right now, without the official Microsoft SDK, developing for the Kinect is hard. Really hard. But if Microsoft really makes an effort with the SDK, and includes all the tools a developer might need--including skeletal mapping to see the position of all of your limbs, voice commands, and gesture recognition--we might see some very impressive apps coming from what's now the Kinect hacker scene. Official developers, we assume, already have access to all of this stuff, but those are big international companies. Microsoft needs to think about the little guy, too.
The bigger problem right now is that the Microsoft SDK is exclusively for Windows 7, which means the developers have to use Windows 7 to make the software, and users will have to use Windows 7 to play the software. We know Microsoft intends the Kinect, or at least the technology behind the Kinect, to be used on computers (and probably cars, phones, refrigerators, and lord knows what else, eventually), but Microsoft will have to expand that to other platforms. The Xbox 360 is obvious, but Apple's Mac OS and iOS platforms, Google's Android, and, to a lesser but no doubt vocal extent, desktop Linux platforms are also important. Microsoft can't limit something as ground-breaking as the Kinect to its own platforms--that's why the iPhone can sync with a Windows PC. Flexibility is important.
I am hugely excited about the Kinect. There hasn't been anything as exciting in all of consumer electronics in years. And I still have hope that Microsoft is just doing some sort of slow-burning strategy that will infuriate in the short-term but eventually pay off. E3 is coming up in the first week of June--the first E3 since the Kinect's release. I have every expectation (eep. Sorry for the pressure.) that Microsoft will show us some stuff that'll blow my socks off, then require me to go retrieve them so my feet don't get sweaty while I'm playing whatever full-body game they come up with. But for right now, my Kinect is dusty. Our collective Kinect is dusty. And I wish it wasn't.
Five amazing, clean technologies that will set us free, in this month's energy-focused issue. Also: how to build a better bomb detector, the robotic toys that are raising your children, a human catapult, the world's smallest arcade, and much more.