I’m here at a Microsoft event where we’re getting into the nitty gritty details of Microsoft’s touch-screen table, Surface. If you want to see an earlier version in motion, check out Matt’s hands-on from CES where he makes a fool of himself. Meanwhile, we’ll be learning about what the thing is made up of, its parts and history (it used to be called “Playtable”), and what the plans are for the future. I got to play around with one for quite some time last night and I’m with Matt and the others in thinking this thing is really incredibly fun. Click on the link below for rest of the info and lots of pictures, and as more events occur I’ll keep you informed.
More Surface stuff here if you’re interested.
So, what I’ve learned so far:
There have been about 7 versions of the Surface so far. They’ve had two development paths: one as a proof-of-concept to make the money flow to the project, and one “real” version with significantly different hardware. The main design problem has been that they need empty space for the light to travel through uninterrupted, both to and from the touchscreen. Originally, their design (as you can see in the pictures) involved a single camera set up with a mirror, which worked well enough to impress the exploratory committee or whatever, but was not nearly good enough for the public. So they moved to their alternative, an array of five cameras. They’re all IR, of course – and the engineer kept mentioning “uncommon manufacturing techniques” and said they were actually in on the parts at the ground floor.
The material of the screen is a lot like a rear-projection TV – it’s essentially precision frosted glass, and they had to tweak it for the surface — they were evasive on just how. The projected image has to be bright to your eye, requiring frostiness, but also let the cameras have a clear picture of objects touching it, which requires transparency. Obviously they’ve found a good middle ground.
I have to be honest that I was a little disappointed when I learned that it doesn’t really recognize objects for what they are magically — phones, paintbrushes, cameras, and so on are not recognized by their shape, but they can be identified by special tags attached to the bottom as you see in the picture. These tags can be simple or complex depending on how specific you need them to be (just a dial? or an individual’s unique ID?) Of course, it would be pretty impossible to just recognize a phone’s model and everything just from its shape, but it would be cool.
There have just been a couple interesting but extremely meta presentations on the inspiration behind the UI and what they’re going for. They’ve tested it out on kids and grandmas and I think they’ve listened to the feedback. I’ve talked with a lot of the people on the team now and they really like what they’re doing and one of the things they all say is that the ideas for applications, games, and all that kind of thing just flow naturally from everyone who gets near the thing. It’s so intuitive, so fun, and it works so well that people instantly have a brainwave one way or the other. They’ve had issues making the UI work well because there’s a lot of interference from the GUI mindset. Users aren’t sure whether to leave Windows behind entirely or bring it with them, so it has to be determined by the designers beyond question. They’ve done a great job with that.
If you haven’t gotten your hands on one of these things yet, do so as soon as you can. The team and I agree that it’s an experiential thing. Once you sit down with it for a minute, there’s an instant explosion in your brain and you see all the possibilities, especially when you see what they’ve already done with it. Their mega-high-res photo zooming app, Seadragon, was extremely impressive and it’s pretty much just a tech demo. More on the way.
Heh. This very pretty presenter (yes, we’re very professional here at CG) set up fly-on-the-wall cameras to watch how people with no previous experience with the thing interact with it and go through the discovery process. These people are going nuts, I wish I had video of it. This one lady is acting like a 3-year-old, it’s hilarious. It’s really that fun. They figure it out really fast and while they’re kind of mystified by what exactly this thing is, but they become experts right away. Some old guy is deciding he’s going to teach people how to make things happen. A little kid is navigating the system like she’s an old hand at it. The social aspect is new to them and everyone reacts differently (which side do they go to? how many people can touch it at once?). Fortunately, they’re being secretly recorded so all the feedback makes it to the dev team.
I didn’t have my video camera with me, but SlashGear did. I promised I’d link. This is a renegade post.