That's sonic feedback not tactile (or visual) feedback. There's an absolutely huge difference.
Rest assured, tactile feedback keyboards are far, far more efficient and accurate than keyboards without such feedback. It's been well studied for a few decades and it isn't even a controversial subject. Pretty much all HCI and human factors people are in agreement on this.
That isn't to say that there's not a place for touch pad interfaces. Rather, they are simply a poor tool for general purpose data input.
I would be surprised if there isn't currently existing patent-work regarding the usage of transducers or other types of electromechanical parts behind touchscreens for the purpose of providing tactile feedback on a touchscreen.
And aside from it being expensive, the kapton flex circuits aren't rated for more than one-million flexes, and after ten million, I can promise they will fail. I don't know if that is a ton of keystrokes or not, but I do know that I built a machine to test flex circuits, and if we ran it constant, it would take about a week to do one million flexes. It went a little faster than one flex per second. Anyway, I'm not convinced that in regular use this keyboard will last longer than a year.
I don't contest any of your claims other than maybe that of key life sufficiency. I don't think there are a lot of computer keyboards rated for much more than that, I found one Logitech where they say 10M, but for the rest that I looked at, they didn't say, so I would expect 1M, which I think has been a standard for some time.
A machine designed to test electromechanical endurance doesn't tell us the habits of a typical user
A machine designed to test electromechanical endurance doesn't tell us the habits of a typical user
This one did -- that was the whole point. To boot, we correlated the data to "real world" tests. Our circuits were 1 - 2mm thick traces of silver conductive ink on PET laminate. These are rolled copper, much smaller pitch, and the bending radius is quite small (check these pictures). It's really hard to say what the longevity is, but I can tell you that we've tested some copper circuits that get dicey at only 35,000 flexes. I'd say it's probably 70/30 that Art did his homework here.
I actually just saw this teardown because I applied for a job with the company that hosts the blog. How timely! However, those guys haven't given me a call-back yet. Nonetheless, looks like a fun place to work.
This one did -- that was the whole point. To boot, we correlated the data to "real world" tests. Our circuits were 1 - 2mm thick traces of silver conductive ink on PET laminate. These are rolled copper, much smaller pitch, and the bending radius is quite small (check these pictures). It's really hard to say what the longevity is, but I can tell you that we've tested some copper circuits that get dicey at only 35,000 flexes. I'd say it's probably 70/30 that Art did his homework here.
What about temperature? Did your test bed result in temperature rises in the material that wouldn't occur if there were more time between flexes to allow for heat to dissipate?
Not in the way I was thinking of. Few press the same key on a keyboard continuously, round the clock. If each key on a keyboard lasts 1M presses, and a random key is pressed on average, once per second, for 40 hours a week, it looks like such a keyboard should last nearly 7 years. This is excluding half the keys, assuming that most people don't use half the keys. Some speed demons will probably kill the switches sooner, but I don't think a whole lot of people type that much.
This was just talking about the switch, you didn't mention the other parts of this keyboard that might go bad.
What about temperature? Did your test bed result in temperature rises in the material that wouldn't occur if there were more time between flexes to allow for heat to dissipate?
Didn't go fast enough to generate substantial heat. Enough of the devils advocate about a machine that you have not much knowledge about, or for that matter a test product you aren't likely familiar with. I don't mean to seem haughty, but I don't really want to explain everything -- it's a lot of information. It is a good machine that produces data that correlates with the usage profile of the target product.
In response to Jeff, moving parts are likely to ail before solid state electronics. All this thing really is: some OLEDs, circuits, and switches. The circuits aren't going to have problems unless they we designed incorrectly. The OLEDs, well yes, they might fail. The flex circuit is the next things I'm suspicious about. If I'm going to lay down a lot of money for a keyboard, I don't want there to be a statistically small but still notable percentage of the product that fails at 35,000 keystrokes, or perhaps less. That's my point, as clear as I can make it. By soing something as simple as re-orienting the way the flex cable connects with the switch, Art Lebedev could all but erase the chance of flex cable failure during the device's lifetime.
Didn't go fast enough to generate substantial heat. Enough of the devils advocate about a machine that you have not much knowledge about, or for that matter a test product you aren't likely familiar with.
Dude, it was just a question, and IMHO, a reasonable one. I'm sure plenty of accelerated lifetime tests end up being not all that realistic due to heat issues. I'm happy to believe you when you say it wasn't an issue in this case, but since you didn't mention it yourself I thought it was worth asking.
This is the kind of thing I was thinking of. The keyboard wouldn't look like that. It would be much neater and possibly removable, but I couldn't be bothered spending much time on this one! I really saw the keyboard as being something of an extension to the screen which has the tactile response that we currently know, but offers much more in the way of configuration. Perhaps just a gel overlay or a pressed shape at the bottom of the actual screen.
Multi-touch is a technology invented for portable use. The advantage of a mouse interface (in non-portables) is that you can move your cursor on a 30" screen from the lower left to the upper right by only moving your hand from one edge of your mouse pad to the other.
In a portable, you don't have a lot of room for the larger interfaces like a mouse and keyboard. And THAT is where multi-touch is useful.
And when you have the precision of a mouse, why do you need gestures? With a mouse, I can easily click and drag the corner of a photo with a slight twitch of the hand, whereas with multi-touch, I'd have to lift both arms and make a huge pinching motions.
I know you are all excited for the iPhone, but cool it on the multi-touch. Think about why we don't have the click wheel as a computer interface for a moment and it might dawn on you why we don't need the multi-touch over what we have now.
Next person who mentions a multi-touch anything other than an iPhone... I'll multi-touch your mum. (JK)
I hate to tell you this, but you are wrong on every count and display woeful ignorance of MultiTouch. Fingerworks created MultiTouch for desktop machines years before Apple bought them out. Your "huge pinching motions" statement makes absolutely no sense and tells me you know nothing about Fingerworks products. I use pinching motions all the time with my iGesture pad to cut to the clipboard. It takes two fingers, like picking up a sheet of paper. Zoom, on the iGesture as well as the iPhone, takes five fingers on ONE hand. Why do I need gestures? Because I have about 30 separate commands at my fingertips without having to reach for the keyboard. And the combination of USB Overdrive and Menu Master reassigns those commands every time I switch applications. For instance, the moment I switch to any of my web browsers, the iGesture is programmed to give me forward, back, home, reload, stop, command-click to open in new tab, open new tab, next tab and previous tab, on top of the regular global functions that I keep the same to provide consistency, like cut, copy, paste, zoom in, zoom out, single click, double click, scrolling by page, scrolling by line, open file, new window, close window, quit application, etc. Now THAT's useful. I have all the same precision as any mouse plus better ergonomics. You may want to learn more about the history and functionality of MultiTouch before screaming that it's only good for the iPhone.
Multi-touch is a technology invented for portable use. The advantage of a mouse interface (in non-portables) is that you can move your cursor on a 30" screen from the lower left to the upper right by only moving your hand from one edge of your mouse pad to the other.
In a portable, you don't have a lot of room for the larger interfaces like a mouse and keyboard. And THAT is where multi-touch is useful.
And when you have the precision of a mouse, why do you need gestures? With a mouse, I can easily click and drag the corner of a photo with a slight twitch of the hand, whereas with multi-touch, I'd have to lift both arms and make a huge pinching motions.
I know you are all excited for the iPhone, but cool it on the multi-touch. Think about why we don't have the click wheel as a computer interface for a moment and it might dawn on you why we don't need the multi-touch over what we have now.
Next person who mentions a multi-touch anything other than an iPhone... I'll multi-touch your mum. (JK)
A GUI, a GUI? Who needs a GUI when you've got a command prompt?
Comments
That's sonic feedback not tactile (or visual) feedback. There's an absolutely huge difference.
Rest assured, tactile feedback keyboards are far, far more efficient and accurate than keyboards without such feedback. It's been well studied for a few decades and it isn't even a controversial subject. Pretty much all HCI and human factors people are in agreement on this.
That isn't to say that there's not a place for touch pad interfaces. Rather, they are simply a poor tool for general purpose data input.
I would be surprised if there isn't currently existing patent-work regarding the usage of transducers or other types of electromechanical parts behind touchscreens for the purpose of providing tactile feedback on a touchscreen.
And aside from it being expensive, the kapton flex circuits aren't rated for more than one-million flexes, and after ten million, I can promise they will fail. I don't know if that is a ton of keystrokes or not, but I do know that I built a machine to test flex circuits, and if we ran it constant, it would take about a week to do one million flexes. It went a little faster than one flex per second. Anyway, I'm not convinced that in regular use this keyboard will last longer than a year.
I don't contest any of your claims other than maybe that of key life sufficiency. I don't think there are a lot of computer keyboards rated for much more than that, I found one Logitech where they say 10M, but for the rest that I looked at, they didn't say, so I would expect 1M, which I think has been a standard for some time.
A machine designed to test electromechanical endurance doesn't tell us the habits of a typical user
A machine designed to test electromechanical endurance doesn't tell us the habits of a typical user
This one did -- that was the whole point. To boot, we correlated the data to "real world" tests. Our circuits were 1 - 2mm thick traces of silver conductive ink on PET laminate. These are rolled copper, much smaller pitch, and the bending radius is quite small (check these pictures). It's really hard to say what the longevity is, but I can tell you that we've tested some copper circuits that get dicey at only 35,000 flexes. I'd say it's probably 70/30 that Art did his homework here.
I actually just saw this teardown because I applied for a job with the company that hosts the blog. How timely! However, those guys haven't given me a call-back yet.
This one did -- that was the whole point. To boot, we correlated the data to "real world" tests. Our circuits were 1 - 2mm thick traces of silver conductive ink on PET laminate. These are rolled copper, much smaller pitch, and the bending radius is quite small (check these pictures). It's really hard to say what the longevity is, but I can tell you that we've tested some copper circuits that get dicey at only 35,000 flexes. I'd say it's probably 70/30 that Art did his homework here.
What about temperature? Did your test bed result in temperature rises in the material that wouldn't occur if there were more time between flexes to allow for heat to dissipate?
This one did -- that was the whole point.
Not in the way I was thinking of. Few press the same key on a keyboard continuously, round the clock. If each key on a keyboard lasts 1M presses, and a random key is pressed on average, once per second, for 40 hours a week, it looks like such a keyboard should last nearly 7 years. This is excluding half the keys, assuming that most people don't use half the keys. Some speed demons will probably kill the switches sooner, but I don't think a whole lot of people type that much.
This was just talking about the switch, you didn't mention the other parts of this keyboard that might go bad.
What about temperature? Did your test bed result in temperature rises in the material that wouldn't occur if there were more time between flexes to allow for heat to dissipate?
Didn't go fast enough to generate substantial heat. Enough of the devils advocate about a machine that you have not much knowledge about, or for that matter a test product you aren't likely familiar with. I don't mean to seem haughty, but I don't really want to explain everything -- it's a lot of information. It is a good machine that produces data that correlates with the usage profile of the target product.
In response to Jeff, moving parts are likely to ail before solid state electronics. All this thing really is: some OLEDs, circuits, and switches. The circuits aren't going to have problems unless they we designed incorrectly. The OLEDs, well yes, they might fail. The flex circuit is the next things I'm suspicious about. If I'm going to lay down a lot of money for a keyboard, I don't want there to be a statistically small but still notable percentage of the product that fails at 35,000 keystrokes, or perhaps less. That's my point, as clear as I can make it. By soing something as simple as re-orienting the way the flex cable connects with the switch, Art Lebedev could all but erase the chance of flex cable failure during the device's lifetime.
Didn't go fast enough to generate substantial heat. Enough of the devils advocate about a machine that you have not much knowledge about, or for that matter a test product you aren't likely familiar with.
Dude, it was just a question, and IMHO, a reasonable one. I'm sure plenty of accelerated lifetime tests end up being not all that realistic due to heat issues. I'm happy to believe you when you say it wasn't an issue in this case, but since you didn't mention it yourself I thought it was worth asking.
"Multi-touch this!"
"Multi-touch that!"
"Ohh - I bet we could Multi-touch that!"
"Yes! A multi-touch toilet flusher! Amazing!"
Multi-touch is a technology invented for portable use. The advantage of a mouse interface (in non-portables) is that you can move your cursor on a 30" screen from the lower left to the upper right by only moving your hand from one edge of your mouse pad to the other.
In a portable, you don't have a lot of room for the larger interfaces like a mouse and keyboard. And THAT is where multi-touch is useful.
And when you have the precision of a mouse, why do you need gestures? With a mouse, I can easily click and drag the corner of a photo with a slight twitch of the hand, whereas with multi-touch, I'd have to lift both arms and make a huge pinching motions.
I know you are all excited for the iPhone, but cool it on the multi-touch. Think about why we don't have the click wheel as a computer interface for a moment and it might dawn on you why we don't need the multi-touch over what we have now.
Next person who mentions a multi-touch anything other than an iPhone... I'll multi-touch your mum. (JK)
Stop it with all of this multi-touch jabber.
"Multi-touch this!"
"Multi-touch that!"
"Ohh - I bet we could Multi-touch that!"
"Yes! A multi-touch toilet flusher! Amazing!"
Multi-touch is a technology invented for portable use. The advantage of a mouse interface (in non-portables) is that you can move your cursor on a 30" screen from the lower left to the upper right by only moving your hand from one edge of your mouse pad to the other.
In a portable, you don't have a lot of room for the larger interfaces like a mouse and keyboard. And THAT is where multi-touch is useful.
And when you have the precision of a mouse, why do you need gestures? With a mouse, I can easily click and drag the corner of a photo with a slight twitch of the hand, whereas with multi-touch, I'd have to lift both arms and make a huge pinching motions.
I know you are all excited for the iPhone, but cool it on the multi-touch. Think about why we don't have the click wheel as a computer interface for a moment and it might dawn on you why we don't need the multi-touch over what we have now.
Next person who mentions a multi-touch anything other than an iPhone... I'll multi-touch your mum. (JK)
A GUI, a GUI? Who needs a GUI when you've got a command prompt?