Regarding the leap...while it has not had as much adoption as the team had hoped, the effort being put into it is still significant. The latest SDK update addresses a number of shortcomings that the initial implementation had, making the hardware that much more viable.
That all said, the biggest issue with all of these new interactions is the lack of a solid, low fail interaction model that can seamlessly replace our current interfaces.
Looking at the leap, while it does a number of things right it has to always get it right, or as close to always as possible to be as good as a mouse click or a keyboard key press.
This looks interesting, but watching a promotional video doesn't show me how often a person has to flick their hand or swipe the touch strip to get the desired result. I have a leap and while I've gotten fairly proficient with it, I still find it's failure rate to be to high to make it a real replacement for my mouse and keyboard.
The problem with Leap is that to do anything remotely accurately for a moderate period of time takes an huge amount of effort moving your arm in mid-air with only visual feedback. It will be marginally worse than using a wii controller to type on a screen keyboard, as that includes tactile vibration feedback as you hover over keyboard letters. In the video it struggled at typing a single word on a keyboard.
The device looks great for toy/demo cases like google earth and fruit ninja. There aren't many normal use cases where you would need slow, low accuracy pointing or gestures from a distance that isn't already covered by wii, kinect or a laser pointer for example.
Remote control is already solved by remote controls with buttons and smartphones connected to a tv. The tactile feedback increases the speed of usage so much that no touchless device will ever surpass.
It has the same problem the Leap has: It has intolerably high latency between movement and execution of input. (Judging from the video in some cases almost a whole second, but generally over half a second.) It might be useful for niche applications, but if i need to take over ten times longer to do simple tasks with it, it will never be part of normal daily usage.
We have made a lot of progress since the early product videos. We are demoing it in SF at the AA hackathon today if you want to see it in action. Nod is the only pixel accurate gesture control device in the market.
I'm in Germany, so seeing it isn't remotely an option. I'd appreciate the hell out of videos though. Youtube is cool and modern phones have really good cameras. ;)
Douglas Adams saw this coming. From the Hitchhiker's Guide to the Galaxy:
For years radios had been operated by means of pressing buttons and turning dials; then as the technology became more sophisticated the controls were made touch-sensitive – you merely had to brush the panels with your fingers; now all you had to do was wave your hand in the general direction of the components and hope. It saved a lot of muscular expenditure, of course, but meant that you had to sit infuriatingly still if you wanted to keep listening to the same program.
Thalmic Labs' Myo is more conceptually similar to this than the Leap. This looks cool but it does look laggy and inaccurate. The Myo alpha unit I got to test worked beautifully.
Hi there - good chance we're either getting blocked by a spam filter or have an incorrect email. We've been sending out some kind of email roughly every 4-6 weeks. If you send us a note at thalmic @ thalmic.com we should be able to get that sorted out.
Wish they had more technical details instead of breathless marketing. Looks like a gyro mouse, as a ring. Definitely neat, but hardly revolutionary. Gyro mice have been around for years (http://www.amazon.com/s/?field-keywords=gyro%20mouse ); they're rather frustrating to use. The ring doesn't look small enough to wear all day, and the latency looks pretty terrible in the demo video.
What I really want more than another gesture thingy is a one-handed, portable, and eyes-free text entry device; something that could really replace a keyboard for a mobile device, even for demanding activities like programming. Is anyone working on that?
I cant find the product right now but I recently saw a bluetooth chorded keyboard. Unlike the soft keyboards chorded keyboards have the added benefit that you can still maintain typing speed and use proper punctuation.
I'd like to see a list of sensors and output devices that it has (from the video I assume it has an accelerometer, gyroscope, 2 buttons, and a touch-sensitive strip, anything else? magnetometer? microphone? light sensor? speaker? vibration motor? LED?), ideally with precision/accuracy, sampling rate, and latency figures. Weight, battery life, wireless range and technology would be nice too.
Ah, after poking around a bit I found wireless range and battery life (though "1 day of active use" isn't very specific). These kinds of specs should be front and center in a list on your developer pages.
github.com/openspatial already documents the wire protocol, has sample code, SDKs for iOS/Android and an emulator.
We invested our time in making sure the SDK was robust and documented well for developers. We plan to add more technical documentation in the future. Stay tuned.
How does this make my life better/easier? Tv's have been trying to do this and nobody really cares. Take the Nest example ... why don't I just literally turn the Nest device?
I'm not saying this couldn't have a practical use -- just that you aren't proving it.
you can point at your Smart TV and actually type into your TV. Ofcourse this requires the smart TV to incorporate our gestural keyboard (and we are working with leading vendors to incorporate it or at least provide it as an app)
I'll stick with a wireless mouse and keyboard for now. The device in the video looked incredibly painful to use. Strange how the producers did not see that.
I just checked out their product. I'm an engineer and it is one of the best wearable devices I've seen out there. The initial product demo video doesn't do any justice to the latest version they have.
I'm excited for these type of wearables as input devices for head-mounted HUDs/AR. Google Glass currently uses voice recognition and a capacitive strip, which is fine, as Glass is only for briefly looking up information. As AR develops, it will need more complex input methods, and I see these as some vague possibilities:
Neat idea but this looks like it will suffer the same annoyances as the leap. It works great when you make very specific controlled movements, but that ends up taking away from what touch-less is supposed to promise in my mind. I just want to flip my hands about casually but the leap would rarely catch it properly. Instead I was usually just left frustrated and tired of holding my arm up in the air.
The part of the video where they traced a word out on the keyboard just looked painful trying to keep your hand up in the air and steady.
Haven't used my leap after the first month of getting it. I fear the nod will be the same.
Do like the button/slider feature however. That seems like a good addition. But that gets me thinking is a wii controller much different?
I actually thought the whole thing was a parody video (of other similar devices) during the clip where a bunch of scribbling across the keyboard somehow ended up spelling "Island". Even the logo makes it look like the device is really hard to control.
Touchless is kind of opposite of what I want. Haptics are important.
In almost every situation I'd rather have a physical knob or button. If I'm sitting down at a desk or on my couch and putting on a ring, I might as well be picking up a controller or remote that have instant action and tactile feedback.
I suppose all these hand waving hardware startups share the Minority Report vision for interfaces.
44 comments
[ 3.3 ms ] story [ 112 ms ] threadThat all said, the biggest issue with all of these new interactions is the lack of a solid, low fail interaction model that can seamlessly replace our current interfaces.
Looking at the leap, while it does a number of things right it has to always get it right, or as close to always as possible to be as good as a mouse click or a keyboard key press.
This looks interesting, but watching a promotional video doesn't show me how often a person has to flick their hand or swipe the touch strip to get the desired result. I have a leap and while I've gotten fairly proficient with it, I still find it's failure rate to be to high to make it a real replacement for my mouse and keyboard.
The device looks great for toy/demo cases like google earth and fruit ninja. There aren't many normal use cases where you would need slow, low accuracy pointing or gestures from a distance that isn't already covered by wii, kinect or a laser pointer for example.
Remote control is already solved by remote controls with buttons and smartphones connected to a tv. The tactile feedback increases the speed of usage so much that no touchless device will ever surpass.
What exactly do you mean with pixel-accurate?
That said, touchscreens still feel less satisfying than physical buttons, so maybe I'm just being nostalgic.
Or just pragmatic :)
There are dozens of articles/papers/essays I could put you, but here's one: http://worrydream.com/ABriefRantOnTheFutureOfInteractionDesi...
For years radios had been operated by means of pressing buttons and turning dials; then as the technology became more sophisticated the controls were made touch-sensitive – you merely had to brush the panels with your fingers; now all you had to do was wave your hand in the general direction of the components and hope. It saved a lot of muscular expenditure, of course, but meant that you had to sit infuriatingly still if you wanted to keep listening to the same program.
What I really want more than another gesture thingy is a one-handed, portable, and eyes-free text entry device; something that could really replace a keyboard for a mobile device, even for demanding activities like programming. Is anyone working on that?
As for programming -- a language could be invented that is on-screen keyboard friendly (no brackets, etc).
https://en.wikipedia.org/wiki/Chorded_keyboard
I've never used one, but I want to. They're so expensive, though :(
Ah, after poking around a bit I found wireless range and battery life (though "1 day of active use" isn't very specific). These kinds of specs should be front and center in a list on your developer pages.
We invested our time in making sure the SDK was robust and documented well for developers. We plan to add more technical documentation in the future. Stay tuned.
I'm not saying this couldn't have a practical use -- just that you aren't proving it.
IMO This is not a solution to anything.
It was a complete pain in the ass to wave your arm around to do everything.
* Hand/wrist/arm mounted input (Nod, Myo, watches, chorded keyboards, etc...)
* Eye-tracking (technology exists, but needs to be miniaturized and developed further to fit within a glasses frame)
* Head movement (tiling your head, winking, etc.)
* Computer Vision (ie. AR develops enough to use your hand in front of you to intuitively interact)
The part of the video where they traced a word out on the keyboard just looked painful trying to keep your hand up in the air and steady.
Haven't used my leap after the first month of getting it. I fear the nod will be the same.
Do like the button/slider feature however. That seems like a good addition. But that gets me thinking is a wii controller much different?
Cool tech though.
In almost every situation I'd rather have a physical knob or button. If I'm sitting down at a desk or on my couch and putting on a ring, I might as well be picking up a controller or remote that have instant action and tactile feedback.
I suppose all these hand waving hardware startups share the Minority Report vision for interfaces.