Project Natal, Microsoft's motion-detecting controller-less interface, was one of the big stories coming out of this year's E3. I'm looking forward to seeing what developers can do with it, but I'm also pretty skeptical about the impact it will actually have on the Xbox 360, or gaming in general.
The E3 promo video for Project Natal reminds me of Apple's Knowledge Navigator from 1987.
In Apple's stark vision of the future, the day will come when each of us will own a "computer book" with our own tiny Bill Nye the Science Guy. Bill understands the context and nuance of casual conversation, and longs for our touch.
Like Project Natal, there's a lots about the Apple video that immediately stinks of too good to be true. For instance, in the future our appliances can be as douchey as we are.
Granted, a lot of the Knowledge Navigator functionality actually has come to pass through google, wikipedia, and newer information gizmos like Wolfram Alpha, but not in the form of a fussy little man in a bowtie who understands conversational English.
Similarly I imagine we'll see a number of ideas from the Project Natal video developed on the Xbox 360, just not in the form we see in the video.
What Didn't Wow Me
The girls shopping via video conference is strange and awkward BS. It seems to suggest that Microsoft will have partnerships with online retailers that allow you to try out 3D versions of their clothes on a realistic 3D avatar of yourself, and allow you to do so collaboratively in a video chat.
I don't see that actually happening. For one thing, I don't know what clothing retailer would want to sink time and money into an Xbox Live fashion marketplace. For another, it's a bizarre and artificial social experience. People go shopping together because it's something to do while hanging out. People don't video chat while shopping on Amazon.com and drop stuff into each others carts for the lulz of it.
It seems like the mandate for the E3 video was "something for the whole family" and they needed some feature to fill the teenage daughter demographic. So they manufactured this world where teenage girls like to go on Xbox LIVE (gold) and put imaginary clothes on waxy rotating mannequins of themselves. It's like a Judy Jetson concept of what girls would like to do with technology.
I don't buy it when the character speaks the player's name in the Kung Fu game or 1 vs. 100. I don't think text to speech is advanced enough yet, and there's no way you could cover every possible name with a voice actor.
In terms of actual hardware functionality, I expect that the Natal camera will be, at best, an EyeToy with depth perception and greater fidelity.
For some types of gameplay, like the Ricochet demo (basically Super Glove Ball but using the full body), an EyeToy with depth perception should be just fine.
The paint splat demo is also something you could do with an EyeToy-like interface. Nothing revolutionary, but it could be fun to dick around with.
I wonder what happens when your cat walks in front of the camera, or mom comes between you and the TV vacuuming. Does the game come to a halt and then need to re-calibrate? What if you do an unintended gesture like scratch your nose while miming the steering wheel. Does the car fly off the road and burst into flames?
Milo & Molyneuxvian Promises
In the Milo demo you can see that milo prompts the user to do a very specific gesture (putting on a mask) at a very specific time.
I'm sure the system will be able to accommodate controlled situations like this very well. In this case pretty much any movement around your face at the right time could be interpreted as putting on a mask.
But, If the player pretended to put on a mask at the wrong time, like when they're over by the tree, I doubt Milo would have interpreted it correctly. He probably would think she was being attacked by bees or may assume that she was being troubled by restless ghosts.
I don't really believe the conversation that the woman has with Milo at the beginning of the video. His use of the name Claire, his very specific reaction to "thousands of people", it all feels scripted. That's fine to communicate the vision, but I don't expect it in the final product.
Basic head tracking and facial recognition seem plausible. One of the more advanced features Milo is supposed to have is the ability to recognize the emotion in your face and reflect it back at you. I'm interested to see how accurate this ends up being in practice. What if you've just got a sour puss? Is Milo always in a bad mood?
Players will probably forgive a lot of wonkiness in these interactions since they don't know what to expect. I suspect a lot of Milo's reactions to facial expression and tone of voice are just going to be wild guesses that are generic enough to cover just about any scenario.
Notice how after the (scripted) conversation Milo changes the subject to catching fish and drawing in his journal. Nothing that he says after changing the subject is at all dependent on the first part of the demo.
Later, the player shows Milo a drawing of fish. Milo's reaction to the drawing is to comment on the fact that it's orange - not the fact that it's a fish. If the user had drawn a smiling orange penis, I bet they would have gotten the same reaction.
I don't think these limited interactions are necessarily a fault of the hardware. Rather I think its a credit to the designers' ability to maintain the illusion in spite of technical limitations.
Ultimately I expect a lot of the the Natal experiences will come down to designing an experience that maintains the illusion of control without actually providing much.
Rambles and Mutters
When it comes to Natal design I imagine that coming up with systems to disregard unintended or confusing input will be just as important, if not more so, than coming up with systems to interpret what you're doing.
If you're sitting on the couch with friends, the system needs to know how to disregard the non-player hands or to ignore someone reaching out in front of the player really quickly to grab the remote.
The driving game made me think that Natal is going to have to be very forgiving in its gesture recognition. Just try miming a steering wheel. Odds are with nothing to hold your hands don't rotate in a nice clean circle. They also probably don't stay evenly spaced at ten o'clock and two o'clock. Natal would need to be pretty lenient in how it interprets the virtual wheel. This makes me wonder how precise it could actually be. Games will have to give credit for being "close enough" if games are going to be any fun.
I expect Natal will be pretty good at at navigating the NXE dashboard. I wouldn't be surprised if the new dashboard was actually designed with Natal's swipe interface in mind.
I also wouldn't be surprised if the buzzer system for 1 vs. 100 already exists in some form. Still, I wonder how good it will be at distinguishing between eight hands smooshed close together.
While it's being billed as a controller-less system, I wouldn't be surprised if MS or third parties offered peripherals. Some might not have any functionality at all. For example a simple hoop of plastic you hold to make miming a steering wheel feel more natural (just like the Wii Wheel). Other's may have reflectors or color coding to help improve the fidelity of the motion detection. I doubt we'll see any Natal controllers with actual buttons or analog sticks, but I suppose it's possible.
Shake Your Baby Maker
Another minor gripe is the name. The word natal is pretty common. For me it conjures up images of floating fetuses, ultrasound jelly, outie belly buttons popping out of veiny stretched abdomens, water breaking all over the gym at the prom, and premature flipper babies baking in incubators.
I realize it is pronounced Nah-Tahll, like first part of Natalia, But unless you've heard someone explain it that way, I think alot of Americans would pronounce it Nay-Tuhl.
Yeah, I know that "Wii" had similar pronunciation confusion and associations with bodily function when it was first announced, but I think this is worse. When I hear, "Microsoft's Project Natal" I think of a clumsy robotic nursemaid who crashes to a blue screen of death or red rings as it's reaching into the soggy vulvoid chamber of secrets.
I suggest pronouncing it Project Gnat-Uhl, in honor of Natalie Portman. Here is a music video from a band called Natalie Portman's shaved head. It features Shamu, the world's deadliest friendliest sea mammal.
Ultimately I expect that Project Natal will provide fun new game experiences. It will probably extend the life cycle of the Xbox 360 a bit, which is a good thing. I expect that like the Wii there will be some very well designed games that do a great job of working with the system's limitations to deliver a great illusion of control. I also expect there will be a lot of crap that overreaches, trying to force the system to do more than it's capable of. And then there will be the Eye-Toy, In The Movies, style mini-games that lend themselves well to minimal controls.
In any event, I welcome Project Natalie Portman without expecting much more from it than some neat shit.
What do you expect from Project Natal? Have you seen anything in the previews that really gets you psyched or makes you super nervous?